A new survey from LinkedIn has shed light on employee’s scramble to be seen as AI-savvy, as the emerging technology takes hold and becomes more common place at work.

Among the skills we’re adding to our profiles, ‘Question Answering’, ‘Classification’ and ‘Computer Vision’ were all among the top AI-related skills mentioned.

We explain just what these mean, and why having AI skills is seen as vital to employers in 2023.

Top Five Growing AI-related Skills

In its recent survey into AI in the workplace, published this week, LinkedIn identified the five fastest growing AI-related skills that users are adding to their profiles.

The most popular AI skills are listed as:

1. Question Answering +332%
2. Classifcation +43%
3. Recommender Systems +40%
4. Computer Vision +32%
5. Natural Language Processing +19%

If these terms don’t mean much to you, then don’t panic too much, there’s still time to catch up. Maybe it’s time to sign up for a free AI training course?

Question Answering is an AI model that can, as you might expect, receive a question from a user in human language, interrogate a database, and respond with the correct answer.

Classification in AI relates to how models group and sort data, anything from recognising the difference between a spam email and a legitimate one, to much more complex sorting, such identifying hate speech one a social media platform.

Recommender Systems, as the name suggests can be used to make recommendations to people. One of the most common, everyday examples you’ll see of this is on Amazon, where data about individuals browsing and shopping habits are fed into the algorithm, and used to suggest products that the user may be interested in.

Computer Vision is the process by which AI makes sense of images, video, audio and so on, and extracts relevant information from them. You have probably come across this when searching your image gallery for a certain photo. Search for the term ‘cat’, and you’ll be presented with all the snaps you’ve taken of furry felines, as AI can identify in the images.

Natural Language Processing, or NLP, is one of the core components of AI platforms such as ChatGPT. It’s the ability to understand prompts and commands written in human language.

AI Skills as a Recruitment Tool

While AI in the workplace is still a relatively young concept, the effect it has had on companies in that short space of time can’t be understated.

In its survey, LinkedIn consulted with executives, and found that AI is very much at the forefront of their minds. 44% stated that they intend to increase their use of AI in their organizations in the next year. 47% also agreed that using generative AI would increase productivity.

However, despite the negative predictions earlier in the year about huge job losses as a result of AI, even from Open AI CEO, Sam Altman, only 4% of those surveyed claimed they would reassess roles and reduce headcount as a result of AI.

AI’s Impact Will Vary Depending on Industry

In its study, LinkedIn notes that while the scope for AI and what it can help workers achieve is huge, the effects won’t be universal.

For example, LinkedIn gives on example of a teacher. It estimates that 45% of a teacher’s role could be complemented with generative AI, through tasks such as lesson planning, training and curriculum development. However, it also argues that more than half a teachers time is spent working with students, which an AI is unlikely to be able to replicate.

Conversely, LinkedIn also gives the example of software engineer roles, which it argues could be augmented up to 96% with AI. With the vast majority of coding time removed, engineers could spent more time developing new skills and learning new systems.

LinkedIn’s picture of AI in the workplace is generally positive. Employers want it, employees can learn it, and it will afford everyone more time. If you’ve got any AI skills in your back pocket, we’d suggest adding them to your LinkedIn profile, and if you’ve yet to jump into the world of AI, check out a free course.

While the heated debate about working remotely rages on, a new workplace disruptor is slowly emerging – the 4-day week.

Thanks to several large trials in the US, UK and Canada, it’s a movement that’s picking up speed, and some US states have already committed to cutting a day out of the working week.

We take a look at which states are switching to a 4-day week, and what it could mean for you.

The Rise of the 4-Day Work Week

The 5-day work week is commonplace today, but it wasn’t until Henry Ford implemented it in his factories in 1926 that it gained traction, being codified into law in 1940. Now, nearly 100 years since Ford decided to shorten the work week for his employees, we’re on the cusp of moving to a 4-day work week.

While the concept isn’t new, there’s certainly never been more conversation about it, as well as detailed research into the benefits of working a four day week. The results have been overwhelming, to say the least.

In a recent Canadian trial, researchers found that revenue increased by 15%, and all 41 companies involved decided to continue with the approach after the trial had ended.

In a separate global study, companies reported that employees were happier, less burned out, and it was easier to attract new talent.

This explains why many companies have already switched to the 4-day week, including Shopify, Kickstarter and Basecamp.

Even Amazon and Microsoft have been experimenting with the shorter work week.

US States With, or Proposing, a 4-Day Work Week

• California
• Massachusetts
• Missouri
• Pennsylvania
• Texas
• Vermont…maybe?

California

In July 2023, Rep. Mark Takano (D-Calif.) put forward legislation that argued for a 32 hour week for Californians.

“I am introducing this legislation to reduce the standard workweek to 32 hours because – now more than ever – people continue to work longer hours while their pay remains stagnant.” – Rep Takano

Under the legislation, companies would need to compensate employees who worked over the proposed 32 hour threshold, under the Fair Labor Standards Act (FLSA).

The bill is being reintroduced by Takano after an earlier attempt in 2021.

Massachusetts

In April 2023, Rep. Josh Cutler (D-Duxbury) and Rep. Dylan Fernandes (D-Falmouth) filed legislation for a 4-day week pilot across Massachusetts, that would last for two years.

The voluntary program is intended to allow companies to offer employees a 4-day week, with no reduction in pay, in exchange for tax credit.

“This bill creates new incentives for Massachusetts businesses to explore shifting to a four-day work week which can offer a myriad of benefits, including boosting worker satisfaction and productivity, and reducing absenteeism and commuting time.” – Rep. Cutler

Missouri

Missourians have a good shot of getting a 4-day work week…but only if they’re a teacher. In August, the state announced that it was switching to a 4-day week for a third of its schools, affecting around 98,000 students.

The move has been made primarily with staff in mind, with 62% of teachers in the state leaving by their fifth year. It is hoped that the move will increase teacher retention, and some schools in the area have already reported a rise in the number of applications, as a direct result of the 4-day week.

The move is one that is being mirrored across other states in the US, in an attempt to retain staff.

Pennsylvania

Democratic Rep. G. Roni Green stated in August that she is looking to introduce 4-day week legislation in Pennsylvania, to enable workers more downtime, and allow them to focus on mental and physical health.

Green’s planned legislation would see businesses with more than 500 employees reduce work hours  from 40 to 32, without a reduction in staff pay.

Small businesses would be unaffected.

Texas

Like Missouri, Texas has been adopting a 4-day week for its students, if not the general working populace. Currently around 60 school districts in Texas operate a 4-day week, with similar logic to Missouri in that it aids teacher retention.

However, not everyone is onboard. In April, Republican senator Donna Campbell put forward a bill to mandate a 5-day week for Texas schools, stating “students cannot afford to lose any more precious hours of instructional time”. The bill drew considerable criticism from teacher’s groups.

Vermont…Maybe?

Before any Vermonters get too excited, we should say that the state has not made an official move to a 4-day week yet, but it’s hard to ignore Vermont senator’s, Bernie Sanders, passion for the concept.

He penned a 2023 opinion piece for the Guardian, where he rallied against corporate greed, and pushed for a 32 hour week. In the article, Sanders states that since the introduction of the 40-hour work week under the Fair Labor Standards Act in 1940, US workers have become 480% more productive, and that a shorter working week is long overdue.

“It’s time that working families were able to take advantage of the increased productivity that new technologies provide so that they can enjoy more leisure time, family time, educational and cultural opportunities – and less stress.” – Senator Bernie Sanders

Given Sanders’ passion for the 4-day week, it’s not to hard to imagine Vermont taking steps toward the 4-day week at some point in the near future.

Is a 4-day workweek coming to your state? Learn how to implement one here.

States That Have Failed to Introduce a 4-Day Work Week

• Hawaii
• Maryland
• Utah
• Washington

Hawaii

In 2022, Hawaii explored the prospect of a 4-day work week for state employees.

In the bill put forward by lawmakers, references are made to not only the mental and physical health benefits, but also the environmental impact, citing a diminished number of Friday commuters.

The proposal was not approved.

Maryland

In January 2023, senators in Maryland proposed the ‘Four Day Work Week Act of 2023,’ designed to incentivize private and public employers to trial a reduced day week, and provide up to $750,000 in tax credits.

In March the proposal was pulled after concerns that it would not pass due to the associated costs, and concerns about ‘institutionalizing’ the 32-hour week.

However, the bills sponsor, Del. Vaughn Stewart, has stated that he intends to try again in 2024, after sourcing more research.

Utah

Way back in 2008, Utah conducted a 4-day week experiment, although the terms were slightly different to the bills we see proposed today.

Instead of cutting from a 40-hour week to a 32-hour one, the pilot saw employees squeeze the 40-hour week into four 10 hour days.

The pilot ran until 2011, when it was closed due to not saving as much money as originally hoped.

Washington

Back in 2020, senator Joe Nguyen proposed a 4-day week for Washington. Under bill 6516, employees would be entitled to work four days instead of five, and those that worked over their 32 hours would be entitled to overtime, at a rate of at least 1.5 times their hourly rate.

The bill was very similar to those being proposed in 2023 by the likes of Maryland and California, although unfortunately, in this case, it didn’t go anywhere.

4-Day Work Week Benefits

The four-day work week is possibly the most exciting way companies are trying to boost productivity and attract top talent in 2024. With promising results in some trials this could mean a win not just for wellbeing, but also for workplace inclusivity and diversity. Four-day weeks could allow more parents and carers to secure full-time jobs that work for them, and input on business strategy.

Given that we know a more diverse workforce is a more successful workforce, we certainly hope to see more trials and success stories across the 50 states, very soon.

Forget being productive. The truly important thing, as any retail middle manager will tell you, is to look productive.

Now, a new survey of nine different countries has logged the time spent at work doing performative work and compared it against the time spent on real work.

The result: A look under the hood at which countries support a workplace culture that emphasizes surface-level busy-ness rather than one about getting down to business.

India, Japan, and Singapore Prioritize Performative Work

Workers in India self-reported the most time spent on performative work (an average of 43%) rather than productive work (57% of their time).

Data from the survey, which polled more than 18,000 desk workers across February and March of this year, recently debuted in Slack’s State of Work 2023 report. 

The two most productive countries were US and South Korea, which both reported the same ratio of time wasted vs time well-spent: 28% to 72%.

What’s Keeping Us From Peak Productivity?

Slack’s report didn’t stop there. Respondents who said their productivity has been dropping listed four different main reasons why they weren’t getting as much done. They might:

• Experience too many unnecessary meetings and emails (70% say that fewer meetings and emails would improve their productivity)
• Feel unsupported by their managers and disconnected from their colleagues
• Lack access to effective tools and technology
• Grapple with information silos and switching between disconnected apps

Connecting with your employees and helping them cut down on meetings might go a long way towards boosting real productivity.

And, since meetings are the biggest issue, reducing unneeded meetings with the right web conferencing software could be a huge help as well.

Is a “Return to the Office” Encouraging Performative Work?

Remote work thrived when the Covid-19 pandemic hit in 2020, but bosses have been pushing for a return to the office every since.

In 2023, that retrenchment towards physical office jobs is in full swing, with the list of companies that have reversed their flexible remote work policies packed with big names, from Starbucks and General Motors to Disney, Walmart, Amazon, and Meta.

But has it helped productivity? Last year, we reported that US productivity rates had fallen by the sharpest rate since 1947, following a record growth in 2021.

And earlier this month, Amazon’s SVP of Amazon Video and Studios Mike Hopkins said he had “no data either way” to justify the company’s in-office work mandate. Even more recently, a study this week found that 80% of employers say they regret rushing the return-to-office push.

Perhaps remote work can help emphasize actual results, and help the US continue to resist the siren call for performative work.

Job listings that don’t mention any degree requirements have reached a massive 90% total, one new HR-related survey has found.

That’s good news for anyone that wants a fair playing field for a skills-based job position — although it might be a bit of a bummer for anyone with a little or a lot of student debt.

Formal educations have long served as barriers for many, and as an easy way for recruiters to cull their stack of applicants. Now, the job economy is moving in a different direction.

75% of Recruiting Professionals Say Skills-First Hiring Is on the Rise

LinkedIn issued the survey that uncovered the shift away from formal education signifiers.

The social platform’s Future of Recruiting report 2023 found that skills-based hiring is on the rise, and could be the future of job placement around the globe.

Not only has the number of job ads in the UK that don’t mention a need for a degree jumped up 90% year over year in 2022, but the number of US job listings that don’t require a four-year degree hopped up to 20% from around 15% during the same period.

“Since 2019, the share of recruiter searches on LinkedIn that include a skills filter has grown by 25% — and today, recruiters are 50% more likely to search by skills than they are to search by years of experience.” ~LinkedIn

Businesses Must to Know What Skills Their Workers Actually Need

Part of a shift towards focusing on skills over certifications? Businesses need to make sure they know just what those skills are.

“Part of having a skills-first strategy is going beyond just hiring to ensure you’re looking at the larger picture. First, you need to understand the skills that your business needs. Then understand what skills your employees have, or that they can develop, and what skills you need to hire for so that you build your talent strategies accordingly.” ~Jennifer Shappley, VP, Global Talent Acquisition at LinkedIn

That’s easy to agree with — a full 94% of recruiting professionals did just that as part of the recent survey.

Not quite as many (84%) say that working towards understanding employee skills is “currently a priority” at their own company, however, suggesting that there’s a little more work to be done on this front.

A Healthy Job Market Might Be Pushing Recruiters Towards the Skills-Based Search

Tech jobs are still being slashed left and right, following a trend that first kicked off just about a year ago, but employment is actually pretty high in the overall economy.

That strong job market might even be the impetus behind the trend away from degree requirements: Recruiters need to fill job positions, so they’ve dropped some of the barriers that had been filtering out perfectly good workers.

As a result, some workers are finding jobs that would have otherwise passed them over. You don’t need a degree to see the economic benefits to taking a skills-based approach to job searching.

We already know that artificial intelligence is set to take thousands of jobs. But is AI about to disrupt your morning commute as well?

A new free-standing traffic camera powered by AI has spotted and photographed almost 300 violations in the first three days of use.

The camera, which is used in the UK as part of a trial program, caught 117 people on their phones while driving, as well as 180 instances of people failing to use their seatbelts while driving.

What to Know About the New Traffic Cam Trial

The new camera’s very first trial run took place recently in south west England, where road safety tech firm Acusensus teamed up with Vision Zero South West.

They installed it in one of the busiest roads in the area, and made sure that every offense was double-checked for accuracy by a human reviewer, according to one report.

The same team had tested a different setup with a larger camera for a 15-day trial last year, but the new system is billed as a free-standing option which can be moved around easily to different roads in any given area.

This new traffic camera is only the latest example of AI innovations impacting everyone’s daily lives. In this case, we might all need to start keeping seat belts buckled and our phones in our pockets when behind the wheel — unless we’re ready to start paying a mind-boggling amount of traffic tickets.

Will AI Traffic Cameras Really Start Ticketing Me?

But there are more than a few speed bumps that AI cameras will need to get over before that actually becomes a reality, if it ever does.

First, it’s just a trial, and hasn’t yet locked in a larger roll-out within the UK. In addition, the trial took place in the UK, so there’s no immediate reason to assume it will expand to the US.

At the same time, we already live in a world in which plenty of our private data is already scraped and can easily be used by the police to crack down on infractions, whether because the police department is unlawfully retaining data or because companies like Facebook offer it up on a platter. Adding AI cameras that can watch everyone at all times might just be dramatically increasing all that private data that’s already being abused.

AI dash cams are definitely helping commercial fleets today, but that doesn’t mean an execess of AI cameras is just as good.

AI Chatbots Have Already Been Fighting Traffic Tickets for Years

Ironically, there’s already an AI tool designed to help the average joe fight traffic tickets: We covered DoNotPay back in 2016, when the “robot lawyer” successfully got around 160,000 parking tickets thrown out in London and New York.

Theoretically, we can trigger an AI battle in which the automated cameras issues tickets and the automated lawyers refute them.

In reality, we might be headed towards a more relaxed version of that highly automated future. We may seen more AI traffic cameras and may have more AI options to help address them. Concerns about a more highly regulated distopian future may well be warranted, however, and if you feel strongly enough about avoiding a world in which an automated camera surveils you from every single street corner, make some noise. ChatGPT might be able to do the first draft.

Google Chat, formerly known as “Google Hangouts”, is a chat-based messaging app that gives businesses and regular users a space to collaborate and catch up for free. The app is wildly popular, boasting over 10 million downloads on Google Play alone. But as with other platforms like LinkedIn, its large, predominantly professional user base also makes it a magnet for cyber scammers.

Unlike messaging apps like Signal, Google Chat’s security reputation isn’t great. Conversations on the platform are largely unmonitored and accounts can be reached by anyone with a Gmail email address –  making users more vulnerable to unwanted advances. Falling prey to scams on Google Chat isn’t inevitable, though.

We’ve spoken to real businesses that have fallen victim to Google Chat scams to save you from making the same mistakes. From your standard-issue phishing attacks to less typical cases of sextortion, here are six scams to be aware of on Google Chat, and advice on how to avoid them.

What are Google Chat Scams?

Google Chat scams refer to any type of cybercrime that takes place on the Workspace platform. Scams take a variety of forms: some grifters send chat requests to users directly, while others try and move conversations to Google Chat from social media sites to take advantage of the app’s lackluster security settings.

Due to the platform’s overwhelmingly professional clientele, Google Chat scams tend to mirror cons spotted on other sites like LinkedIn. But cons aren’t just focused around the nine till five, with romance and sextortion scams also becoming more prevalent on the app in recent years.

Here’s our summary of the most common scams currently circulating Google Chat:

• Google Chat Phishing Scams
• Google Chat Technical Support Scam
• Google Chat Fake Job Offers
• Google Chat Romance Scams
• Google Chat Sextortion Scams
• Google Chat Bogus Giveaways
• Google Chat Crypto Scams

Google Chat Phishing Scams

Unsurprisingly, the most popular scam to be aware of on Google Chat is phishing – a social engineering practice where criminals masque as reputable sources in an attempt to lure sensitive information from victims.

Due to its low barrier to entry and relatively high success rate, phishing scams are now the most common type of cyber attack in the US, and Google Chat’s public access makes it one of the easiest platforms for phishers to target.

Nikita Sherbina from the SaaS company Aiscreen.io almost fell victim to the hustle after receiving a seemingly innocuous message that claimed to be from a well-known client.

“The message contained a link that appeared to be a shared document, but upon closer inspection, it led to a fraudulent login page” Sherbina tells Tech.co. Fortunately, her IT department was quick to shut the request down, but not all Google Chat users are this luckily.

A phishing scam spotted at the University of Miami. Source: miamioh.eduDaily

Google Chat user Rhett Stubbendeck from LeverageRx fell victim prey to the ruse after receiving a “seemingly innocent notification appeared from a very trusted financial institution”.

“What happened afterwards was very unsettling”, he tells Tech.co. “My business accounts were compromised, which gave access to unauthorized parties and gave way to potential data breaches.” 

How to avoid this scam: While the rise of generative AI is making phishing requests increasingly hard to decipher from the real deal, telltale signs include an inflated sense of urgency, requests for private information, and misspelled or suspicious-looking email addresses.

Learn more about some red flags to look out for in our guide to avoiding phishing emails.

Google Chat Technical Support Scam

Another common tactic plaguing Google Chat is IT support scams.

Also known as technical support scams, this type of fraud involves fake IT teams reaching out to users and claiming they’ve found a problem with their device. The scammer will then urge the target to install applications to safeguard their systems, or click on a link to resolve the problem.

Humayun Saleem, a marketing consultant for Clashify, told us about a time when his colleague was targeted by this type of scam.

“The scammer claimed their computer was compromised and offered assistance for a fee” Saleem told us. Luckily, his teammate realized it was a scam before any damage was done. 

“My colleague wisely ignored the message, highlighting the importance of skepticism and verifying the legitimacy of such claims.”

Similarly to LinkedIn, this scam is pervasive on Google Chat as the platform primarily caters to professional communities. There are a number of tell-tale signs that indicate requests may be illegitimate, though.

How to avoid this scam: Legitimate tech platforms like Google Chat won’t contact you on the app to notify you about errors on your account.

Real IT teams rarely ask you to click on links to resolve issues, either. So if you’re notified about a technical issue on your account try and reach out to service providers directly before following through with any requests.

Google Chat Fake Job Offers

As layoffs sweep across the US, cybercriminals are using fake interview scams to exploit willing candidates looking for work.

Typically, scammers will do a little background research before setting up the honey trap. This may involve logging your employment history on LinkedIn or googling your name to see what sector you’re currently working in.

Fake job interview taking place on Google Chat. Source: reddit.com

It’s common for fake recruiters to reach out to you via LinkedIn or email first, before setting up an “interview” on Google Chat where they’ll continue the charade and try to entice you about the proposed job offer. The perpetrator will then request some type of fee in order for you to move forwards with the process.

Liam Lucas, CEO and Founder of the digital tech company Off Road Genius told us about a time he nearly fell prey to a fake job offer scam.

“I’ve encountered fake job offers where the scammer promises a high-paying job in return for a fee” Lucas shared. In his case, the fee that was demanded was “purportedly for background checks or training materials”, but luckily he declined the bogus offer before any hard-earned cash was exchanged. 

How to avoid this scam: Real recruiters will never demand payments before moving you along the onboarding process. Be wary of any company that makes such requests, and do your background research before getting excited about a new job opportunity.

Google Chat Romance Scams

While Google Chat may be an unlikely place to find love, the platform is a popular destination for romance scams, a type of crime where grifters feign a romantic interest with targets before attempting to extort them.

Romance scams are typically orchestrated by international criminal groups who scope out victims on dating sites like Tinder or Bumble. After establishing a rapport with targets, bad actors will try and move the courting process over to platforms with weaker security protocols like Google Chat.

According to Australian psychologist Monica Whitty, perpetrators will then try to develop hyperpersonal relationships with the victim until they feel like they’re ready to depart with their money.

Romance scammer posing as a military doctor. Source dailymail.co.uk

Due to the scam’s reliance on emotional manipulation and deception, Whitty explains the ruse often causes a “double-hit” among victims, as those targeted are left with financial damages alongside the abrupt loss of a personal relationship.

Unfortunately for susceptible singletons, instances of romance scams have shot up 91% since Covid-19, with women making up two-thirds of reported victims.

How to avoid this scam: If someone you meet online asks to move the chat over to third-party messaging apps like Google Chat, immediately question their movies. Also, don’t share financial information or any other personal details before meeting up in person.

Google Chat Sextortion Scams

Sextortion is a type of blackmail where criminals threaten to publish private, sexual content if victims don’t meet their demands.

While cases of sextortion are on the rise across most platforms, Google Chat’s current settings – which don’t allow users to delete chat history for both parties – make it the platform particularly well-suited to scammers looking to engage in this type of crime.

Like romance scams, the particularly distressing type of scam is normally established on dating platforms before being moved to apps like Google Chat. Then, after both parties are using the platform, the scammers will work on building trust, before eventually requesting compromising images, videos, or information from the target.

But it’s not just those handing over sensitive materials that are at risk. It’s becoming increasingly common for sextortion scammers to create sexually explicit content from benign images of victims using photo editors and AI technology – opening the scam up to a much wider pool of potential prey.

How to avoid this scam: Be weary of user that suddenly steer the conversation towards sexual themes and refrain from sharing explicit content with unverified accounts.

If you suspect someone is blackmailing you on Google Chat, document all potential evidence, instantly report the suspect to the platform, and contact the authorities.

Google Chat Bogus Giveaways

Everyone loves a freebie. But with this type of online crime, the only party getting a reward is the scammers.

Google Chat giveaway scams take place when cybercriminals reach out to users on the platform, claiming that they’ve won a payout, product, or service. The grifter will pretend to be a credible organization and ask targets to hand over financial information in order for the prize to be released.

The scam can take many forms, but Google Lotto and Google Pop-up scams are especially common on the platform. Both types of fraud target Google users and ask them to complete a survey or enter personal information to claim their winnings.

Google Chat prize scam. Source: support.google.com

Not all bogus giveaways claim affiliation with Google, though.  Tina Grant, tech manager at aerospheres, told us about a personal experience a relative had with this type of scam.

“They eventually got him to share is credit card details for “security verification” and to pay for “shipping”. They completed a $2,800 transaction on his card.” – Tina Grant, Tech Manager at aerospheres

“(My dad) was told he won an Apple hamper from his bank and being someone who always wanted to use Apple but never really “got to it”, he decided to reply” Grant revealed. Unfortunately for her father, he parted with $2,800 before realizing the prize was a hoax. 

How to avoid this scam: Unfortunately, if a giveaway seems too good to be true, it probably is. If you haven’t entered a competition or sweepstakes, any claim that you’ve won a prize should be a giant red flag – especially if they’re requesting personal information.

Google Chat Crypto Scams

While our understanding is that crypto scams aren’t as common on Google Chat as they are on other platforms – and they’re certainly less prevalent than tech support or fake giveaway scams referenced above – they’re still worth watching out for.

Cryptocurrency has been providing criminals with the perfect cover for their scams for several years now, as they take advantage of the many legitimate (but arguably still questionable) investment tip-offs and opportunities lining our social media and email inboxes, along with the ones we see pushed by trading influencers online on a near-daily basis.

In Google Chat crypto scams – as well as crypto investment cons taking place on other platforms – lucrative investment opportunities with near 100% returns and next to no risk are often advertised as a simple way to get rich quickly. The way the value of cryptocurrencies like Bitcoin has grown over time, as well as the hype around new, exciting digital currencies like Ethereum, make it strangely easy for cyber criminals to convince the average punter that their tip-off is genuine.

How to avoid this scam: The ol’ if it’s too good to be true, it probably is line is worth trotting out here for a reality check. If an investment opportunity really is low-risk and can guarantee a purported high reward, why is it being farmed out to people’s message, social media, and email inboxes, and not front page headlines? No one should be making any investments off the back of unsolicited text messages.

Is Google Chat Safe?

Google Chat offers several built-in security features like end-to-end encryption and automatic phishing detection, giving users a decent guarantee of protection. Like many messaging apps, the provider uses Transport Layer Security (TLS) to encrypt messages sent between devices and its servers.

However, Google Chat’s low barrier to entry and external domain messaging options make it a lot less scam-proof than other messaging platforms like Signal and Telegram. This doesn’t mean you’re destined to be duped, though. By practicing due diligence, avoiding suspicious-looking links, and keeping personal information to yourself, you stand a solid chance of remaining safe on the app.

It’s also worth knowing how Google uses your data when you use Google Chat. On a help and support page, the tech giant explains that “features like Smart Reply and spam detection, as well as a smarter product experience” can only be provided if the company can access and use chat data.

“To improve Chat’s performance and reliability,” and to help with troubleshooting in case of issues while you use Chat” Google continues, “we collect performance data and crash analytics. We also use this info to help prevent abuse of our services and for analysis.”

This is all pretty standard stuff, and Google also says on the same page that Google Chat data isn’t used for advertising. However, the company can and will access your chats in response to a legal order from the government – so bear this in mind when using the platform.

Protecting Yourself from Online Scams

In 2024, many of the online scams you’ll see lurking about are specifically tailored to the platforms they’re being spread on. In other words, cyber criminals know that generic correspondence won’t work. This means scams are constantly changing, and often, the ones you receive via email look very different from the ones you get sent over SMS or Google Chat. This makes them pretty tough to spot at times.

However, there are still some rules to follow to avoid all phishing scams. Firstly, we’d avoid clicking any links you receive via text message purporting to link you to a legitimate company’s page. Supposedly missed a “delivery”? Log in with your reference number on your email address. Are your accounts being “deleted”? Contact the company directly. Remember, you can always open a separate line of communication with any company if something feels off.

Additionally, treat email SMS, and other correspondence trying to inject a sense of urgency into your decision-making with extreme caution. While legitimate companies like the often-impersonated PayPal might want you to “sign up now” for a limited-time offer, that’s very different from a “representative” of a company suggesting your account will be suspended immediately if you don’t hand over personal information, or pay a fee.

Another thing you can do to protect yourself is to ensure you’re never reusing passwords across accounts. Although this is more of a form of damage limitation for when your credentials for one site are compromised, it’ll ensure the cyber criminal in question is unable to get into any of your other accounts with your information. If you’re reusing the same password over and over again, one breach is all it takes to expose everything.

Google Chat Scams FAQs

With businesses finding new, inventive ways to make money with ChatGPT every day, it’s no surprise that AI training courses are becoming increasingly sought after.

Workers in all sorts of industries are looking to future-proof their jobs and upskill themselves in line with the rapid technological changes occurring. Luckily, companies like Microsoft and Google offer free AI training courses, as do some higher education institutions.

In this guide, we cover the best AI training courses currently available, as well as the benefits of learning about AI in the current job market. We’ve largely focused on free courses that offer immediate, foundational learning opportunities that you can start applying to your job role or career straight away, rather than paid degree courses that cost hundreds or thousands of dollars.

• Best Free AI Training Courses for Generative AI
• Best Free AI Training Courses For Programmers, Developers & Tech Experts
• Free College AI Courses and Training
• The Benefits of Learning About AI
• The Benefits of Developing Your Soft Skills in an Automated World
• AI Training Course FAQs

The 8 Best Free AI Training Courses for Generative AI

Here, we’ve compiled a list of the best free AI training courses that focus on generative AI and how you can harness it, as well as foundational concepts in artificial intelligence. A lot of these courses are designed to be introductory sessions and geared toward beginners.

1. Google’s Generative AI Learning Path (5 Courses)
2. Microsoft’s “Transform Your Business With AI” Course
3. Google’s “Machine Learning Foundational Course”
4. LinkedIn’s “Career Essentials In Generative AI” Training Course
5. “IBM: AI Foundations for Everyone” Training Course
6. Digital Partner’s “Fundamentals of ChatGPT” Training Course
7. DeepLearning.AI’s “AI for Everyone” (Coursera)
8. Phil Ebner’s AI Crash Course (Udemy)

1. Google’s Generative AI Learning Path (5 short Courses)

One of the more generous courses available in terms of actual hours of learning, Google’s Generative AI Learning path has five short courses on it. All courses take one day to complete.

Seven of the courses are classified as introductory, including “Introduction to Generative AI”, “Introduction to Large Language Models” and “Generative AI Fundamentals”.

There are three courses within the learning path described as intermediate, including “Encoder-Decoder Architecture” and “Attention Mechanism: Overview”.

The first two introductory courses cover a lot of immediately applicable content, such as how to use prompt tuning to get the best out of large language models. There’s also a course on the responsible usage of AI.

Although there’s no official qualification, you will be awarded a completion badge that you can attach to your digital resume. Google has also recently expanded its paid certificates program which can be accessed through Coursera and has made its free courses available in 18 additional languages.

2. Microsoft’s “Transform Your Business With AI” course

This Microsoft learning path is designed, as the tech giant says, to help businesspeople acquire “the knowledge and resources to adopt AI in their organizations”, and explores “planning, strategizing, and scaling AI projects in a responsible way.”

Microsoft says the objectives for this course are to become familiar with existing AI tools, understand basic AI terminology and practices, and use prebuilt AI to build intelligent applications.

To enroll in this course, which is 2 hours and 40 minutes long, Microsoft says you’ll need a “basic understanding” of IT and business concepts. Modules included in the pathway are:

• Leverage AI tools and resources for your business (55 mins)
• Create business value from AI (21 min)
• Embrace responsible AI principles and practices (48 mins)
• Scale AI in your organization (36 mins)

3. Google’s “Machine Learning Foundational Course”

Along with the foundational AI course mentioned above, Google also provides an introductory course on machine learning that’s available for free. You can access the course with your Google Workspace or personal account, and there’s actually a surprising amount of material included, considering it’s free.

Included as part of the material is a “crash course” to machine learning, a section on “problem framing” and a course on how to prepare your data for machine learning workflows. Included are freely accessible videos explaining core concepts, provided by machine learning experts. You’ll then be able to “check your understanding” at the end of each section with a quiz.

While this course is a little more advanced than the other Google course to make this list – and has a bit more of a business focus – it’s worth checking out if your company is looking for new ways to implement AI/machine learning in the workplace.

4. LinkedIn’s “Career Essentials In Generative AI” training course

LinkedIn’s AI Career Essentials Course is made up of five different videos, with a total run time of around four hours. Each video is hosted by a different AI expert, covering a range of core concepts and ethical considerations relating to AI models.

One of the videos provides a detailed explanation of how to streamline your work with Microsoft Bing Chat, while another discusses the key differences between search engines and reasoning engines. Although there’s no accreditation or certification, completing this course will earn you a badge of completion from LinkedIn, which can be displayed on your profile.

The fifth video in the series, entitled “an Introduction to Artificial Intelligence” is an hour and a half long and provides a simplified overview of the best AI tools for businesses, which is handy for those who haven’t taken the plunge yet and implemented a tool in work.

5. IBM’s “AI Foundations for Everyone” training course

IBM offers a course entitled “AI Foundations for Everyone” through Coursera, of which over 2,000 people have already signed up for in just February 2024. You can audit the course for free, which will give you access to all of the materials and some of the assignments, but you won’t be graded or get a certificate at the end. For that, you’ll need a Coursera subscription.

It’s geared toward beginners and you don’t need prior experience to enroll, and the schedule is flexible so you can learn at your own pace. Along with AI fundamentals, the course will also ensure you’re familiar with IBM’s own AI services, which help businesses integrate artificial intelligence into their existing infrastructure.

IBM says that, by the end of the course, participants will have had “hands-on interactions with several AI environments and applications”.

The course has three modules: “Introduction to Artificial Intelligence”, “Getting Started with AI using IBM Watson”, and “Building AI-Powered Chatbots Without Programming”. Each module takes between nine and eleven hours to complete.

6. Digital Partner’s “The Fundamentals of ChatGPT” training course

Digital Partner’s course entitled “The fundamentals of ChatGPT” is a great option for anyone who wants to take a free, accredited course that covers the basics of Generative AI. As of February 2024, more than 8,000 learners have completed the short course.

During the course, you’ll spend time learning about OpenAI’s role in global AI development, and be able to learn about how ChatGPT works, its advantages and limitations. There’s also a variety of examples included within the course that will show you how to leverage ChatGPT for different tasks, and you’ll learn more about the difference between ChatGPT and ChatGPT Plus.

Modules include “Working With ChatGPT”, “ChatGPT and Its Shortcomings” and “Training a GPT Model”. This free course is available on alison.com, and is published by a digital marketing firm called Digital Partner. The course is CPD accredited and a certificate will be awarded upon completion of a small assessment at the end of the 1.5-3 hour program.

7. DeepLearning.AI’s “AI for Everyone” (Coursera)

Not to be confused with IBM’s similarly-named online AI training course, DeepLearning.AI’s “AI for Everyone” short course is worth enrolling in just to see what the fuss is all about. 1.2 million people have enrolled in this course and more than 40,000 have reviewed it, with an average score of 4,8/5 in February 2024 – perhaps because it’s run by lecturer Andrew Ng, an instructor with over 40 courses published on the site.

This course is split up into four broad modules: What Is AI?, Building AI Projects, Building AI in Your Company, and AI in Society. All four of the modules take less than an hour to complete, so you could feasibly complete the course in a day if you’re prepared to take the quizzes designed to recap each module.

Like most Coursera courses, you’ll be able to audit the course and access all the materials for free, although you’ll have to sign up for a paid account if you want the relevant accreditation once you complete the course.

8. Phil Ebner’s ChatGPT, Midjourney, Firefly, Bard, DALL-E” AI crash course

While some good courses on Udemy will guide you through the ins and outs of MidJourney and other AI generation tools, Phil Ebner covers the most ground, and more than 30,000 students have already enrolled in the course, which has a 4.6/5 rating on Udemy.

The course is almost two hours long and also includes content that will help you better use tools like ChatGPT to generate text responses as well as images.

The “AI for Visual Creativity” section, however, will show you how to use both MidJourney and Dall-E to create “photorealistic images, illustrations, and digital art in a variety of styles. With OpenAI’s Sora video generator seemingly just around the corner, there’s never been a better time to learn the basic principles of AI video generation and the tools that can do it.

On Udemy, you don’t receive certificates of completion for free courses, but if you’re just looking to upskill yourself free of charge, this course is definitely worth a look.

The 6 Best Free AI Training Courses for Programmers, Developers & Tech Experts

Up next, we have more advanced courses geared towards programming and development.

1. Harvard University’s “Introduction to Artificial Intelligence with Python”
2. DeepLearning.AI’s “ChatGPT Prompt Engineering For Developers” (Coursera)
3. Intro to TensorFlow for Machine Learning (Udacity)
4. Georgia Tech’s Reinforcement Learning (Udacity)
5. Become an AI-Powered Engineer: ChatGPT, GitHub Copilot (Udemy)
6. Great Learning’s “ChatGPT for Beginners” training course

1. Harvard University’s “Introduction to Artificial Intelligence with Python”

Harvard University offers a self-paced, 7-week course on the “concepts and algorithms at the foundation of modern artificial intelligence”.

The time commitment of between 10-30 hours a week – but it’s completely free to enroll and you’ll be supported as you complete projects and attend lectures. However, you need to have taken Harvard’s “Introduction to Computer Science” course first to enroll.

2. DeepLearning.AI’s “ChatGPT Prompt Engineering For Developers” (Coursera)

This course will help you utilize OpenAI’s API to write more effective prompts, learn how large language models can be used to carry out tasks like text transformation and summarizing, and teach you how to program and build a custom AI chatbot.

The course is run by AI expert and DeelLearning.AI co-founder Andrew Ng and OpenAI’s Isa Fulford, and it’s only an hour long. DeepLearning.AI says the course is “free for a limited time”. A basic understanding of Python is needed, but aside from that, it’s beginner-friendly.

3. Intro to TensorFlow for Machine Learning (Udacity)

This course teaches participants how to build deep-learning applications with TensorFlow, one of the most popular open-source Python software libraries.

The estimated completion time for the course is approximately two months, and you should have some experience with Python syntax, including variables, functions, and classes, as well as a grasp of basic algebra.

If you take the course, providers Udacity say, you’ll get “hands-on experience building your own state-of-the-art image classifiers” as well as other types of deep learning models.

4. Georgia Tech’s Reinforcement Learning (Udacity)

This Georgia Tech course is free on Udacity and focuses on exploring “automated decision-making from a computer-science perspective”.

At the end of the course – which takes approximately four months to complete, but is also described as self-paced – participants will recreate a result from a published paper on reinforcement learning.

However, it is recommended you have a graduate-level machine-learning qualification and some prior experience with reinforcement learning from previous studies. Experience with Java is also required.

Although there’s no official certificate awarded for completing the course, you can earn a nano degree program certificate by completing Udacity’s 4-month long “Deep Reinforcement Learning”, although this costs $1116.

5. Become an AI-Powered Engineer: ChatGPT, Github Copilot (Udemy)

In this course, students will learn how to create high-quality pieces of code using ChatGPT and integrate it with other text editors. It also covers how to use GitHub Copilot.

This might be a free tutorial, but the course has much better reviews than some of the other AI courses available on Udemy, with 56% of watchers who left a review giving the course five stars, and a further 24% giving it four stars at the time of writing.

The course will be best suited to developers who want to leverage AI tools for coding responsibilities in general, and also, to become more efficient in their coding practices.

6. GreatLearning’s “ChatGPT for Beginners” training course

This is a completely free, two-hour long beginners-focused ChatGPT course. It’s one of the only beginner’s courses on the internet that includes a section on coding prompts, although it also covers quite a bit of other ground, including email prompting.

There are no prerequisites needed for this course, and it has an average rating of 4.61/5, with 75% of reviewers giving the course 5 stars.

The 6 Best Free College AI Courses and Training

A number of universities and colleges offer AI-focused courses.

1. Stanford University’s “Machine Learning” Course (Udacity)
2. Vanderbilt University’s “Prompt Engineering for ChatGPT” (Coursera)
3. Georgia Tech’s “Machine Learning” Course (Udacity)
4. The Open University’s “AI Matters” Course (OpenLearn)
5. University of Pennsylvania’s “AI For Business” (Coursera)
6. University of Helsinki’s “Elements of AI” and “Ethics of AI” Course

1. Stanford University’s “Introduction to Artificial Intelligence” course (Udacity)

This foundational online program, which takes around 10 months to complete at a rate of 10 hours a week, focuses on fundamental AI concepts and practical machine learning skills but is classified as an intermediate course.

The course, which is split into two umbrella sections (“Fundamentals of AI” and “Applications of AI) is completely free if you sign up for Udacity (which also doesn’t cost anything). Unlike Coursera courses, you have access to the full range of materials and teaching, rather than just the ability to audit. It consists of 22 different lessons and a string of interactive quizzes.

2. Vanderbilt University’s “Prompt Engineering for ChatGPT” (Coursera)

Jules White, Vanderbilt University’s associate dean for strategic learning programs and associate professor of computer science, has launched a free online course available through Coursera focusing on prompt engineering.

It goes through the most effective approaches for prompt engineering, covering summarization, simulation, programming, and other useful ways you can harness the power of ChatGPT with your inputs. It’s got one of the highest approval ratings we’ve seen for an AI training course, scoring 4.8/5 with an approval rating of 98% as of February 2024.

The course takes around 18 hours to complete and is made up of an introduction to prompts and three separate sessions on prompt patterns, as well as a 2-hour module on examples. You can either audit this course for free, which means you get access to all the materials but no proof you’ve completed it, or pay for a Coursera subscription (roughly $59 per month). However, you won’t have to pay anything on top of that fee for this course.

3. Georgia Tech’s “Machine Learning” course (Udacity)

In collaboration with Georgia Tech, Udacity has made an intermediate machine-learning course available for free, which takes around 4 months to complete, although the course listing says you can do it at your own pace.

The course is offered as part of an online master’s degree at Georiga Tech, but taking this course won’t earn you credit toward this degree. It includes information on Supervised and Unsupervised Learning, which are two different types of machine learning, and covers how they’re used in AI systems.

However, having a “strong familiarity with Probability Theory, Linear Algebra, and Statistics” and prior experience with statistics is helpful. Students should also have some experience with programming.

4. The Open University’s “AI Matters” course (OpenLearn)

The Open University is a UK-based institution that offers a free course through its learning portal OpenLearn entitled “AI Matters”.

In the course, you’ll learn about the “historical, social, political and economic issues in AI”, explore the benefits and limitations of the technology, and discuss ethical risks relating to AI.

The course is six hours long, and you’ll be eligible for a “statement of completion” from the organization, which has university status in the UK.

5. University of Pennsylvania’s “AI For Business” (Coursera)

The University of Pennsylvania’s “AI for Business” specialization is made up of four different, free courses:

• AI Applications in People Management
• AI Fundamentals for Non-Data Scientists
• AI Applications in Marketing and Finance
• AI Strategy and Finance

In 2023, the University of Pennsylvania’s website detailed that the course itself costs $39 to complete, and you could enroll in the four individual modules for free. Each module took around two hours a week to complete.

However, in 2024, they’ve upped the price to $79. However, we’d recommend accessing it via Coursera – the modules take around 7-9 hours to complete, and if you purchased the 7-day free trial, it could be done in that time. You can also audit the course for free, which means you won’t have proof of completion.

6. University of Helsinki’s “Elements of AI” and “Ethics of AI”

The University of Helsinki has two, free online courses available. The course entitled “Ethics of AI” is geared towards “anyone who is interested in the ethical aspects of AI”, the university says.

The course will familiarize you with common questions that arise in AI ethics and the various ways to approach them.

“Elements of AI” is a broader course with 6 chapters, focusing on topics such as “neural networks”, “machine learning” and “AI problem solving”.  All you need to do to access the course materials is sign up.

The Benefits of Learning About AI

Of course, completing an AI training course can have several benefits. From a personal learning perspective, it’s one of the best ways you can spend your time – AI is here to stay, and getting a better grasp of how it works might just help you out in the near future.

Plus, the things you learn about AI will be applicable to a wide variety of job roles in almost every sector of the economy, so it’s arguably a safer bet than completing a course on a niche or industry-specific topic.

What’s more, right now, businesses are looking for people who understand how generative AI tools like ChatGPT work, and how to leverage them effectively. Employees that are conscious of the limitations of AI tools and able to generate useful responses using prompts are going to become more sought after than employees without these skills.

Completing an AI training course is going to look good on your CV, which will help if you’re applying for a new job. Evidence that you’ve taken the initiative to explore an emerging technology is definitely something an employer will find desirable.

Of course, if you don’t have much of a budget – or you’re not entirely sure what AI training course would be the best use of your time – then trying out some free options is a great place to start.

The Benefits of Developing Your Soft Skills in an Automated World

If you’re looking for AI training courses to enhance your understanding of something that’s going to dominate the business world for years to come, you’re already one step ahead of the crowd.

Completing an AI course isn’t the only way to prepare yourself for the seismic changes to the global economy that will occur as AI is harnessed by more businesses.

One other thing you can do if you want to upskill yourself is to develop your soft skills. These are the “human” skills that AI tools like ChatGPT don’t have. Businesses are always going to need humans to organize, prioritize, and take responsibility for planning and executing projects.

With AI taking control of more and more repetitive administrative tasks, having strong soft skills is going to be increasingly important to stand out in a pool of highly-qualified job applicants or colleagues jostling for promotion.

Developing your soft skills can consist of taking an accredited course – but they are more than often improved by taking experiences, especially in the workplace. It’s vital, then, to put yourself forward for opportunities and experiences that will let you expand your communication, organization, and general leadership skills. You can be taught to work a piece of software in a week, but you can’t click your fingers and make someone a proficient public speaker or project manager.

FAQs

X, the social media network formerly known as Twitter, has been accused of delaying links to websites that Elon Musk is known to dislike.

In test run this week by the Washington Post, links to websites such as the New York Times, Instagram, and Threads were reportedly delayed by around five seconds.

Although the billionaire and self-proclaimed free speech absolutist is yet to comment on the claims, is it even possible to slow down links in this way? We take a closer look.

What Has Twitter/X Been Accused of Doing, and Who Is Impacted?

The Washington Post has claimed this week that Twitter/X is slowing down the loading speed of links accessed through the platform by up to five seconds, a practice commonly referred to as “throttling.”

However, the Post says that a lot of loading speeds returned to normal shortly after the article was published.

Substack, Facebook, Threads, Blue Sky, Reuters, and the New York Times all reportedly experienced link delays. Elon Musk has, in the past, criticized all of them publicly for a variety of different reasons.

Is It Possible Twitter Is Slowing Down Links to Sites Musk Hates?

In a technical sense, it’s entirely possible that the Washington Post is correct, and that Twitter/X is tinkering with loading times for specific sites.

When you click on a link on Twitter/X, you’re redirected to it through a domain called t.co, which Twitter/X uses to shorten links posted on the platform. This is where the Post says the link delays were originating from.

How quickly or slowly traffic passes through that domain is at the whim of Twitter. The platform will have full control over this. However, it’s important to remember that no one working at Twitter/X has confirmed or denied that this was intentional, despite repeated requests for comment.

But Is it Legal?

There’s no explicit or direct legal obligation for Twitter/X — or any other website for that matter — to ensure that external links load within a specific timeframe.

Whether there’s a duty to treat all links the same, however, and not discriminate against specific companies or organizations, is another question entirely. Such a practice may violate certain consumer protections and specific pieces of antitrust legislation. It’s unclear at the moment how strong the legal arguments for either side would be.

It’s Only a Few Seconds — Is That Even a Big Deal?

Yes. It might only be a few seconds in real terms, but when it comes to user experience and website analytics, that’s a long time.

For instance, when a website’s load time increases from one to three seconds, bounce rates increase by about 32%. If a page takes longer than three seconds to load, Google says, then 53% of users abandon the page altogether.

For sites impacted by the link delays, this could result in a major loss of traffic and subsequent advertising revenue.

Although this is unlikely to impact a site the size of the New York Times — at least in the short term — it certainly doesn’t send positive signals to smaller publications and ecommerce businesses deciding which social media platform to make their home.

The private healthcare data of over four million Colorado citizens has been compromised, after threat actors successfully exploited a vulnerability in the MOVEit transfer app used by IBM.

The attack was made possible after the Russian ransomware gang Clop first exploited the MOVEit vulnerability in June, in a seismic cyberattack that affected hundreds of global organizations including The US Department of Energy, the BBC, and Shell Gas.

The breach, which also impacted Missouri’s Department of Social Services (DSS), occurred just a week after Colorado’s Department of Higher Education fell victim to a similar MOVEit exploit that wiped 16 years’ worth of data from its systems.

Private Healthcare Information Stolen in Latest MOVEit Exploit

MOVEit’s file transfer vulnerability has claimed yet another victim. Colorado’s HCPF has recently notified over four million individuals about a May 2023 breach that led to their personal and health information being compromised.

In HCPF’s data breach notification, the agency explained the data was compromised because one of the state’s main hardware providers IBM “uses the MOVEit application to move HCPF data files in the normal course of business.”

It states that while their systems weren’t exploited directly, “certain HCPF files on the MOVEit application used by IBM were accessed by the unauthorized actor”.

These files stolen by the unauthorized actor contained sensitive information relating to approximately 4,091,794 individuals, including full names, social security numbers (SSNs), Medicaid ID numbers, Medicare ID numbers, dates of birth, clinical health care data, and more.

HPCF has pledged to provide victims with two years of credit monitoring services via Experian to counteract fraud attempts like phishing attacks.

News of IBM’s exploit broke just a week after the Colorado Department of Higher Education (CDHE) announced it experienced a ransomware attack that wiped data dating back to 2004.

Fortunately, no HCPF or Colorado state government data was leaked in either attack — but other recent MOVEit victims haven’t been so lucky.

Medicaid Data Exposed in Missouri

Colorado wasn’t the only state to get tangled up in IBM’s recent security breach. The MOVEit attack also affected Missouri’s Department of Social Services (DSS) – a state agency that also uses IBM as a vendor – although the scale of the impact is currently unknown.

In a recent data breach notification, the state agency explained while “the data vulnerability did not impact any DSS systems” it did impact “data belonging to DSS”. They confirmed that this includes health information belonging to Medicaid participants in Missouri.

“”Upon receiving a security bulletin from Progress, we severed interaction of MOVEit Transfer with the department’s IT systems to avoid any further impact to Missouri citizens and their data. No IBM systems were impacted.” – Missouri’s Department of Social Services

Compromised data included individuals’ names, dates of birth, and medical claims. However, only two social security numbers were exposed and no banking information was breached in the attack.

MOVEit Exploit Claims More Victims

News of these healthcare attacks emerged just two months after the Russian ransomware group Clop threatened to publically expose MOVEit victims that didn’t corporate. The ransomware gang gave the affected organizations until June 21 to pay their ransoms, threatening to publish their private information if demands weren’t met.

A number of high-profile names were spotted on this list, including The US Department of Energy, John Hopkins University, 1st Source Bank, and Shell Gas. However, Clop hasn’t claimed responsibility for the IBM breach, and neither the HCPF nor the DSS has appeared on the gang’s victim list — raising doubt about whether they were involved with these attacks.

But whether or not Clop is behind these latest attacks, one thing is for certain: we can be sure to expect more MOVEit mass exploits in the future.

It’s now become even easier to sift through Amazon reviews, thanks to the retailer’s new feature which uses generative AI to summarise customer-generated product reviews.

The highlight feature, which has been in testing since the start of the year, picks out shared themes and the overall “customer sentiment” of reviews, before summing them up in a short paragraph on the product detail page. But what about fakes?

Amazon claims it will be using machine learning models that analyze “thousands of data points” to prevent phony reviews from appearing in summaries. But with 200 million fake reviews being blocked from the platform in 2022 alone and tools like ChatGPT making their creation easier than ever, it’s unclear whether the company’s efforts will be enough.

Amazon Launches New AI Review Summary Feature

As part of Amazon’s recent effort to incorporate artificial intelligence into more of its products and services, the ecommerce giant has just launched an AI review summary aimed at streamlining the customer experience.

Amazon’s new tool uses generative AI to create review summaries, saving customers the time and effort involved with digging through swathes of user comments manually. The feature draws from the most frequently mentioned features and customer opinions and condenses its findings into a short paragraph on the product detail page.

According to a recent press release by Amazon, this new AI-powered feature aims to “help customers determine at a glance whether a product is right for them.”

Amazon’s new product insights feature. Source: aboutamazon.com

The ecommerce platform has also released a “product insights” feature (pictured above) which lets customers filter reviews by common search categories like “Ease of use” and “Stability” to gain targeted insights about the product they’re considering.

Can These AI Reviews Be Fully Trusted?

Amazon’s new review summary feature gives shoppers an easy way to measure customer sentiments at a glance. But how is the company planning to combat the millions of fake reviews that exsist across the site?

According to Amazon’s recent press release, the company is currently investing significant resources to proactively stop the proliferation of bogus reviews across the platform. These investments are funding “machine learning models that analyze thousands of data points to detect risk” including factors like review history and sign-in activity, as well as filtering conducted by human moderators.

“The new AI-generated review highlights use only our trusted review corpus from verified purchases,” – Vaughn Schermerhorn, director of community shopping at Amazon

Amazon began using AI to track and take down reviews in June, after AI chatbots like ChatGPT contributed to an uptick of forged reviews on the site. Amazon also cracked down on the practice throughout 2022, blocking more than 200 million fake reviews and pursuing legal action against multiple review brokers.

Clearly, the ecommerce behemoth is taking fraud detection seriously. But with an analysis from Fakespot finding that 42% of Amazon’s 720 million reviews were fake in 2020, we’d still recommend taking these AI-generated reviews with a grain of salt.

Want to jump into AI but not sure which tool to use? We compare ChatGPT to rival Bard.

Could AI Change Customer Reviews Forever?

Amazon’s new AI summary feature marks the biggest overhaul of its review process seen since they were first rolled out in 1995. But while Amazon’s review summarization rollout may be the biggest to hit the ecommerce world, it’s not the first.

In May, Microsoft added AI-generated app review summaries to its store, displaying top-line review synopsis’ alongside its usual five-star ratings. Online retail company Newegg also launched AI-generated review summaries just last week, boasting custom lists of pros and cons based on customer feedback.

Now Amazon has made the move, it’s fairly likely that these AI-powered summaries will set an industry standard. And with CEO Andy Jassy recently revealing that every Amazon business is running multiple AI initiatives, customers can expect to see a lot more AI upgrades across the site soon.

ChatGPT may be incredibly popular right now, but a new report found that its creator OpenAI could be headed for bankruptcy as early as 2024.

There’s no denying that ChatGPT and the subsequent generative AI platforms that came after it have transformed business as we know it. From improving productivity to spurring mass layoffs, the technology is poised to make a massive impact on world.

Still, not all revolutionary technology is profitable from the get-go, and OpenAI is going to have to find a way to make more money from ChatGPT if it wants to stay viable.

ChatGPT Is Losing Users Fast

If you aren’t growing, you’re dying. At least, that’s the general consensus in the tech industry, with firms like Meta and Google in a constant battle to add users and improve functionality of their platforms.

Unfortunately, one of the biggest problems that OpenAI is currently facing is the number of users taking advantage of ChatGPT. More specifically, the generative AI platform has been hemorrhaging daily active users. They’ve been on the decline since May, when the total number had reached an impressive 1.9 billion.

Now, at only 1.5 billion, ChatGPT has lost 21% of users in just two months, and the numbers are not trending in the right direction.

ChatGPT Competitors on the Rise

So, why are users leaving ChatGPT? The report from Analytics India Magazine posits that the availability of the API has made it so users can access the large language model (LLM) that powers the platform without actually having to use the comparatively limited ChatGPT platform itself.

However, as is often the case in the tech world, a big problem is that ChatGPT competitors are popping up all over the place. Google Bard, Meta’s Llama 2, and Anthropic’s Claude AI have all rolled out with success, drawing their own users and cutting into the brief monopoly that OpenAI and ChatGPT enjoyed since November 2022.

While ChatGPT and its creator OpenAI have a lucrative partnership with Microsoft, which saw $10 billion in funding, that money is the only thing keeping the company in business, considering it costs $700,000 per day to keep the platform running.

Is ChatGPT Going Anywhere?

To be fair, the tech industry is filled with great ideas that aren’t profitable. The ride-sharing industry, for example, has been transforming the transportation industry for years, but its poster child Uber has only just become profitable this year.

Suffice to say, ChatGPT is a powerful, productivity-driving technology that clearly has a market for use. And with OpenAI backed by the likes of Microsoft, it’s safe to assume that the company isn’t going to declare bankruptcy any time soon.

Still, growth is important no matter who signs your checks, which means ChatGPT needs to turn it around in a big way if it wants to be the go-to option for generative AI technology.

The push to get employees back in the office has backfired in a major way, with 80% of employers stating that they regret rushing the decision in a new study.

The return-to-office movement of 2023 has been a tumultuous one to say the least. From big tech firms to small startups, the decision to get rid of remote and flexible work policies has caused everything from strikes to walkouts at businesses like Apple and Amazon.

Employers are finally starting to get the picture, though, with a new study revealing that the majority of them aren’t happy with their decisions.

Employers Regret Early Return-to-Office Decisions

According to a new study from Envoy, titled Without accurate data, the physical workplace won’t survive, 80% of executives would’ve approached their return-to-office strategy differently if they had access to better workplace data.

“Many companies are realizing they could have been a lot more measured in their approach, rather than making big, bold, very controversial decisions based on executives’ opinions rather than employee data.” – Larry Gadea, CEO and founder of Envoy

The study surveyed 1,000 US businesses and their executives, all of which have employees working at least one day per week in the office.

A lack of meaningful data backing up the return-to-office mandates has been a big problem for employers and employees alike. In fact, an Amazon executive admitted last week that their chaotic return-to-office mandate has backed by “no data either way,” which has fueled the substantial backlash from employees.

The Return-to-Office Backlash

With employees around the world begging for more flexible working accommodations and no data to back up the alternative, it’s understandable that these decision makers regret their return-to-office decisions.

Honestly, after all the widely publicized backlash in response to return-to-office mandates across the business world, business owners and employers should be happy that regret is all they’re feeling.

Amazon employees staged a walkout earlier this year, Apple employees have formed a coalition to address hybrid work policies being removed, and other tech firms are having trouble attracting new recruits due to their new in-office mandates.

One study even found that a large majority of employees would start looking for new jobs if flexible accommodations ended, which means that these kinds of mandates are cutting into your ability to retain top talent.

Statistics About Remote and Hybrid Work

If you want to avoid regretting your decision to force your employees back into the office, your best bet is to actually utilize data-driven insights to make the right call. Fortunately, we’ve collected some work-from-home research that should help you decide whether a return to the office is the right call.

For starters, remote workers experienced a 22% performance boost when asked to work from home. Additionally, 47% of employers noted a productivity increase when switching to a more flexible work policy for employees.

If productivity isn’t your primary focus (which would be ridiculous), remote work also saves businesses money. In fact, one study found that businesses can save $11,000 per remote employee per year.

Simply put, remote work and flexible accommodations are the new norm for businesses that want to compete in a meaningful way. And if you rush the decision without any data, there’s a pretty good chance you’re going to regret it.

After demanding remote workers back at least three days a week in February, Amazon is cracking down on staff attendance, but its approach is leaving a lot of employees confused.

According to emails obtained by the insider, a number of workers have been incorrectly identified as breaching the RTO policy, with some believing the error was caused by a technical bug.

Amazon has been quick to own up to its mistake, but with company employees famously resisting the crackdown and the news breaking just a week after a company executive claimed to have “no data” to back the policy change, this development definitely isn’t a good look for the ecommerce giant.

Amazon Fumbles Return to Office Clampdown

Amazon is doubling down on its return to office mandate, but not very successfully.

This week, the online retail powerhouse sent out a series of emails to employees it believed were not complying with its three-day-a-week office policy.

The email claimed they weren’t meeting the company’s “expectation of joining (their) colleagues in the office at least three days a week” even though their assigned office was ready.

Amazon claimed the notification was sent out to workers they believed hadn’t physically clocked in at least three days a week for at least five of the past eight weeks, when their building had been ready eight weeks or more.

“I’ve seen many people across the company who believe they received the same notification incorrectly.” – Anonymous Amazon employee

Unsurprisingly, Amazon’s accusatory emails raised some eyebrows amongst the workforce. According to internal tickets filed on Wednesday afternoon, falsely identified employees are confused and angered by the company’s actions, with at least one worker believing the error was caused by a technical bug in data generating and processing.

In Elon Musk’s latest legal quandary, X – the rebranded social media company formerly known as Twitter – has been fined $350,000 for failing to hand over Donald Trump’s Twitter data to the Department of Justice (DOJ) on time.

While Musk eventually cooperated with authorities, the court ruling declares that the Executive Chair “initially delayed production of the materials,” after arguing the requests violated his First Amendment rights.

Purposely disrupting the special counsel’s investigation into a former President might represent a new low for serial tech entrepreneur Musk, but it’s not his first eye-roll moment this month. X recently drummed up controversy for blinding local residents with an obnoxious flashing ‘X’ sign on Twitter HQ, before suing a non-profit campaigning against digital hate.

Elon Musk vs the DOJ

Elon Musk’s X-Corp has landed itself in legal water yet again for failing to comply with the Department of Justice over data relating to Trump’s involvement with the 2021 insurrection on the Capitol.

Special counsel member Jack Smith issued a search warrant to Twitter in January, requesting access to Donald Trump’s @realDonaldTrump account.

Smith also asked the company not to disclose this warrant to anyone – including the former President – as this may jeopardize the investigation by giving him an “opportunity to destroy evidence, change patterns of behavior [or] notify confederates.”

Unfortunately for the DOJ, Musk didn’t want to play ball. In an appeal that has since been rejected, Twitter argued the nondisclosure order violated his First Amendment right to freedom of speech. On February 7, US district court judge Beryl Howell found Twitter in contempt of court and gave the company until 5 pm to hand over the account information.

After Musk ignored this request, the social media company was fined $50,000 a day, with the sum doubling each successive day. The South-African billionaire finally handed over the documents three days after the court-ordered deadline, with the total outstanding fee amounting to $350,000.

Is Musk Cozying Up to Donald Trump?

After news of Twitter’s noncompliance broke, Musk has been accused by multiple outlets of delaying the government’s access to evidence on purpose. While X-Corp hasn’t released an official statement on the matter, Musk’s ties to the former President are hardly a secret.

Despite being former rivals of sorts when Trump briefly launched his own Twitter alternative, Musk made headlines last November for reinstating Donald Trump on the platform, alongside equally controversial music artist Kanye West. Both figures had previously been blocked from the app after breaking content moderation rules and Tweeting offensive content.

While Musk and Trump may seem like unlikely allies – with the X-Corp boss describing himself as “neither left nor right” and previously claiming the former President is “too old to be chief executive of anything, let alone the United States of America” – the two public figures are both united against censorship on the platform.

However, whether Musk’s noncompliance was ideological is yet to be determined. The DoJ’s spar with Elon Musk hasn’t been the only controversy X-Corp has been shrouded in this month, though.

X’s Kamakazi Mission Continues

After Twitter’s unexpected rebrand to “X-Corp” late last month, the social media company faced legal action from Meta and Microsoft as both companies already own “X” trademarks. X-Corp also received a fine from San Francisco authorities after affixing a giant flashing X sign on the top of Twitter HQ, and has since been asked to take it down.

More recently, the social platform has been fined by AFP, the world’s oldest news agency, for failing to compensate the publisher for displaying its news on its site. Whether Musk decides to comply with the AFP and make a payout, or continue burying his head in the sand, this chaotic sequence of events suggests that days at the newly branded X-Corp are definitely numbed.

Microsoft has just announced a new Dynamics 365 Field Service Copilot initiative which utilizes generative AI to cut out monotonous tasks burdening the service industry.

Microsoft Copilot was primarily focused on streamlining processes for office-based workers, but this new pivot sees the company leverage its large language model assistant to improve operational efficiency on the front line, helping to “address the challenge of repetitive tasks and burnout” as a result.

Microsoft isn’t forgetting about desk-based workers altogether. The software company also announced it would be launching Microsoft Edge for Business next week, a corporate-focused search engine that aims to be the standard browser experience for organizations.

Microsoft Unveils Copilot for Its Dynamic 365 Field Service

In Microsoft’s latest AI experiment, the Seattle-based company has announced a series of new features that aim to increase the efficiency of frontline processes.

While the tech giant may be more familiar with streamlining desk-based practices, these new tools aim to “help frontline workers thrive” by addressing issues currently plaguing the global service workforce, such as labor shortages, device overload, and disjointed systems. But how exactly does Microsoft plan on improving things for companies such as HVAC, lawn care and plumbing businesses?

One of the company’s biggest new announcements is its new Copilot and Dynamics 365 Field Service integration, which will allow service managers and technicians to automate repetitive tasks like work order and schedule creation, freeing up valuable time as a result.

Copilot new Shifts plugin for Microsoft 365 Copilot. Source: microsoft.com

Building on Microsoft’s original Copilot model, which assists office-based tasks like creating documents, and reading and summarising emails, this field service assistant is targeted at the global 2.7 billion-strong service industry, where cases of burnout and poor retention are at their highest.

“Technology can relieve pressures on the frontline that are causing burnout as well as help organizations drive engagement and a sense of belonging that can help increase retention,” – Charles Lamanna, CVP of business applications at Microsoft

Speaking to VentureBeat, Charles Lamanna, Microsoft’s CVP of business applications, explains this initiative marks the first step the company is taking to infuse next-gen AI and data with productivity tools to “address the challenge of repetitive tasks and burnout” and to relieve the burden for frontline workers.

Microsoft Announces New Field Service Integration with Teams

Microsoft has also released a Dynamics 365 Remote Assist app in Teams mobile, which allows technicians to receive and issue updates and request remote assistance from specialists. This new integration will also make it easier for technicians to view upcoming work orders through their home interface.

“Our new Dynamics 365 Remote Assist app in Teams mobile will enable problem-solving in real time with remote experts using 3D spatial annotations that lock to the physical world.” – Charles Lamanna, CVP of business applications at Microsoft

This new Dynamics 365 Remote Assist app will also harness data from Microsoft Teams’ new Shift plugin, allowing team members to keep an eye on upcoming shifts and gain clearer visibility over their outstanding tasks.

Microsoft Edge for Business Will Be Rolled Out Next Week

It’s been a busy week at Microsoft HQ, as the company is also preparing for the release of Microsoft Edge for Business – a new search engine that aims to be the “standard browser experience for organizations”.

The browser, which was first announced at Microsoft Build in May, will bare a lot of resemblance to the company’s flagship Edge browser. However, as the search engine has been designed with the needs of businesses in mind, it will also feature enterprise-grade security, additional productivity features, and enhanced admin controls, according to this recent blog post.

Microsoft Edge for Business is due to be released alongside Microsoft Edge stable version 116, on the week commencing August 17.

It’s well documented that scammers are everywhere online, but a new report from call security specialist Hiya reminds us not to overlook good old fashioned phone fraud, highlighting the ongoing prevalence of Amazon scams and a recent surge in Venmo-related deception.

According to the firm’s latest Global Call Threat Report for Q2 2023, Amazon continues to inspire a large amount of instances of phone fraud and is the second most common scam call in the UK.

In the US, however, mobile payment service Venmo has leapfrogged Amazon for the last quarter and is the third most prevalent pretense phone scammers operate under, after pretending to be family members and the long-running auto warranty ruse.

Watch Out for Venmo Scams

Usefully, the Hiya report reminds us of how the most common Venmo scams operate. Venmo scams aren’t anything new, of course, and typically they’ll still start with an unexpected call or text message explaining that there’s been an unauthorized charge on your account.

Alternatively, the criminals operating the scam may say there have been suspicious log-in attempts on your account. After that, you’ll be connected to a phony customer service rep, who’ll try and get you to verify your Venmo details with them.

Overall, Hiya data shows that as many as one in every four phone calls received in the US can be classed as spam, with the company reporting a suspected 6.5 billion spam calls worldwide in Q2 2023.

AI and the Future of Fraud

We’ve reported on the rise of AI voice scams before and the new Global Call Threat Report reminds us that artificial intelligence is bolstering the scammer toolbox in new and frequently terrifying ways.

Specifically, it once again highlights the extreme case where AI was used to fake a child’s kidnapping and demand a ransom payment.  Phone fraud is just one of the many AI and ChatGPT scams you need to watch out for these days, though the company warns that AI likely represents the future of phone fraud.

“AI voice-generated scams are very likely the model of the future and we expect to find more evidence going forward,” commented Kush Parikh, President of Hiya.

Wham, Bam, No Thank You Scams

Phone fraud probably isn’t news to you, but the Hiya report underlines the unfortunate fact that you just can’t seem to get away from scammers these days.

As illustrated by the continued popularity of Amazon and Venmo scams, anything related to online shopping and payments continues to be ripe for exploitation by criminals, while even LinkedIn scams are causing a blight on the otherwise staunchly professional social network.

Throw in AI changing the fraud landscape in the same, dramatic way that it’s reshaping other aspects of our lives and it’s clear that cybercriminals have come a long way from the humble email phishing scam, dangerous as these still can be.

Now, we’re not saying that you should treat every unexpected email, text, or phone call with suspicion. We’re just saying that’s what we do.

Slack has unveiled a bold new redesign that the Salesforce owned company hopes will help users improve their focus, collaboration and productivity when using the app.

The update, which will roll out to Slack’s estimated 20 million users over the “coming months,” is fundamentally designed to streamline its most important functionality and make it easier to navigate.

Navigation is something some veteran Slackers argue has gotten overly complicated in the app over the years, with the numerous feature additions often distracting from Slack’s core messaging purpose – and sometimes getting lost in the Slackosphere themselves.

Slack Redesign: 6 Features You’ll Actually Care About

The first thing you’ll notice when taking in the Slack redesign is its clean new UI, which is meant to be both easier on the eye with a slightly softer default purple hue, as well as easier to use.

Then you’ll start to become aware of the various changes that have taken place in the interface, with our six favorite new Slack redesign features being as follows.

1) Dedicated DM Section

The heart of the redesign for us is the new far-left navigation bar added to Slack. This now features just six icons to help you decide how you want to navigate the app, with one of them being dedicated to that most old school of Slack features: the DM, or direct message.

The six choices on the new nav bar are better than half the 13 options I currently have on offer in my pre-redesign Slack sidebar – and suffice to say a quicker way to jump to the stuff that actually matters is a welcome addition to the platform.

Image credit: Slack

2) New “Create” Option

Slack probably does a whole lot more than you realize, it’s just much of its functionality gets lost amidst the dreaded current Channels list.

In the new Slack, you’ll have a central “Create” option that allows you to not only message a channel or send a DM, but start a Huddle video or audio call; use Slack’s unloved document collaboration feature, Canvas; or start a new channel entirely.

3) There’s Now An Activity Feed

Whisper it softly, but Slack has added an Activity Feed in its 2023 redesign that reminds of a similar view offered in rival workplace collaboration platform Microsoft Teams.

It’s a single location where you’ll find all your “@” mentions, threads, reactions, notifications and more, so you’ve got an easy way to see everything that potentially needs your attention in the app.

Image credit: Slack

4) Saving Stuff for Later Gets New Life

Did you know Slack has a special way you can save the messages you want to look at and action later? We didn’t really, either.

Fortunately, that’s changing in the Slack redesign thanks to the “Later” icon taking pride of place on the main sidebar. Displaying this more noticeably should encourage people to actually use the feature, which is intended to help you better prioritize your workload and therefore improve productivity.

5) Easier Switching Between Multiple Workspaces

Not everyone will be important enough to have multiple workspaces to toggle between, but for those who are, Slack’s new Enterprise Grid organizational view is a godsend.

It will let you see all of your channels and messages, from multiple workspaces, in one centralized view via the Home menu, which promises a much slicker user experience for the bigwigs out there.

6) Huddles Added to Every Chat Window

Most people are probably guilty of thinking that video calling at work is a case of Zoom vs Google Meet. There are plenty of other video conferencing apps out there, of course, but it’s these two and maybe a couple of other big players like Teams who tend to dominate the discussion.

Slack hopes this redesign will earn its Huddle video and audio chat feature a seat at the table. Moving forward, every messaging window will have the option to start a Huddle in the top right-hand corner. You still might not use it, but at least now you’ll know it’s there.

Google finds itself at the eye of the AI ethics storm once again this week, as the contentious issues of AI “deepfake” technology and also data scraping have the tech giant battering down the hatches both at home and abroad.

In the US, Google is seeking to agree a deal with record industry superpower Universal Music that compensates artists for the use of their voice and melodies by generative AI platforms, which imitate these to create “deepfake” songs, usually without the original artist’s consent.

Down Under in Australia, Google is in a seemingly more combative mood, arguing in court that the burden should be on publishers to opt-out of data scraping, if they don’t want their content to be used.

I’m A Barbie Girl In An AI World

Google is currently facing two significant AI dilemmas, in seemingly contrasting ways, and on opposite sides of the globe. The issues at hand are the use of AI to repurpose music made by human artists and the use of AI to repurpose human-generated web content.

The more pressing of the two is arguably Google’s negotiations with fellow global conglomerate Universal over a music licensing deal that effectively covers sampling by generative AI platforms.

This was first reported by the Financial Times, which cites four people familiar with the matter, as confirming the discussions are currently at an early stage.

The ultimate goal of these talks, according to the sources, is to develop a tool that both lets fans use AI to create so-called “deepfake” songs for Google-owned YouTube content, but also pays the copyright owner for what is being used.

At present, AI is a bit of a Wild West when it comes to what you can find out. You may well have heard a long-deceased Johnny Cash “covering” Barbie Girl if you spend enough time online, while no doomscrolling session on Instagram is complete until you see Harry Potter/Breaking Bad/The Sopranos reimagined as a 90s sitcom.

Precedent Exists In The Form of YouTube

Unsurprisingly, artists themselves are less than impressed with their voices being used in uncontrollable ways. Big names such as Drake (one of many global megastars, including Taylor Swift and BTS to be represented by Universal) have spoken out against their works being effectively cloned without permission.

In addition to the viral rendition of Barbie Girl, you now also have entire social media outlets dedicated to AI songs, with rappers Tupac and Notorious B.I.G. among the other deceased stars to feature prominently on the @PluggingAI channel

What might a final deal between Universal and Google look like? It’s too early to say, but a precedent of sorts exists in the current agreement between YouTube and the music industry, over the use of copyrighted songs in user-created video. This system, which was hard-fought to begin with, already pays out an estimated $2bn a year to the music industry and its artists, for the use of their songs.

Google Opts Out of Responsibility For Web Scraping

There’s little doubt a deal with Universal to help Google create a legitimate AI music tool would be a major boost for the search giant as it looks navigate largely uncharted technological waters. Rivals such as Microsoft and its search engine, Bing, are likely to benefit from tight integration with ChatGPT, though Google has its own chatbot product in Bard – and the Bard vs ChatGPT debate is another matter entirely.

Less certain is how long it will be permissible for Google and other big AI players to scrape the web as part of training their AI systems. According to Google, the issue does also warrant addressing, though it argues that responsibility rests with publishers to opt-out of have their data and content trawled. That’s per a Guardian report that cites Google’s submission to a review of the AI regulatory framework in Australia.

Apparently, the Mountain View-based firm argues that copyright law should be amended to explicitly allow for web scraping by default, saying that it amounts of “fair use” of publicly published content.

Those who don’t want to have their work included in AI training models should be able to opt-out or otherwise specify how their works can be used, not unlike how Creative Commons licenses currently operate. While what happens in Australia could end up having no bearing on the wider AI and copyright debate, it’s still an interesting one to monitor as AI lawsuits like the one fronted by Sarah Silverman descend on courtrooms in the US and all over the world.

A major new scientific study is looking to turn the popular notion of social media stalwart Facebook being bad for your mental health on its head.

Researchers at the world-renowned University of Oxford’s Internet Institute studied individuals in 72 countries for over 12 years and say they found no evidence linking Facebook usage to poor mental well-being. On the contrary, the report argues it may have discovered “quite the opposite.”

In challenging the popular link between social media platforms like Facebook and psychological distress, the two principal researchers – Professor Andrew K. Przybylski and Professor Matti Vuorre – explain their methodology as relating data surrounding Facebook adoption to responses to benchmark mental well-being questions over time.

Is Facebook Actually Good For Mental Health?

In the full study published on the Royal Society website, the Oxford Internet Institute (OII) researchers reveal that they measured mental health sentiment based on data from the Gallup World Poll Survey. The internationally recognized annual study asks respondents things like, “Did you smile or laugh a lot yesterday?” as well as if they experienced more negatively associated feelings like “worry”, “sadness”, or “stress.”

This was then compared to global user base data provided by Facebook for the years 2008 to 2019, during which adoption of the platform spiked massively. The idea here is that if Facebook usage had a genuinely negative impact of people’s mental health, this would be reflected by a noticeable increase in negative sentiment in the Gallup over the same period.

Instead, Professor Przybylsk says the analysis “indicates Facebook is possibly related to positive well-being” – though he makes clear this an anecdotal suggestion, not empirical evidence.

“We found no evidence suggesting that the global penetration of social media is associated with widespread psychological harm” – Principal Researchers, Vuorre and Przybylski

“This is not to say this is evidence that Facebook is good for the well-being of users. Rather, the best global data does not support the idea that the expansion of social media has a negative global association with well-being across nations and different demographics,” he explains in an OII statement.

The Problem With the OII’s Study

While the study claims to offer the first empirical evidence that Facebook usage is not overtly linked with a decline in mental health, there are a number of caveats worth mentioning here.

First and foremost, the two data sets – while substantial – are separate, and therefore their findings are not inextricably linked. While the study relates reliable information in a compelling manner, it is impossible to prove a “cause and effect” relationship between the two. Given how complex a phenomenon mental health is, it therefore seems unwise to assume any cast iron conclusions from the study.

If you were being hyper-critical, you could say the OII study is notable for what it doesn’t prove, rather than what it does.

In addition, the researchers admit that “the observed associations were small and did not reach a conventional 97.5% one-sided credibility threshold in all cases”, and that “in the United States, 13- to 17-year-olds are more likely to use TikTok, Instagram and Snapchat than Facebook, so the user base of Facebook now consists of relatively more older individuals”.

However, despite these caveats, previous research has aligned in some ways. For example, Stanford University found that deactivating Facebook could negatively impact mental health in some individuals, in 2019. 

Social Media Concerns Aren’t Going Away

It’s impossible to try and judge overall social media trends by a single platform. As much as anything, Facebook is a very different beast to newer platforms like Instagram and TikTok, which are much more visual than Mark Zuckerberg’s social network has ever been.

Visual content has been found by a number of prominent psychological studies, such as the Psychology of Popular Media Culture, to result in a higher level of toxic self-comparisons and the pursuit of unrealistic ideals, in turn leading to social anxiety and self-image mental health issues.

It’s one reason many people find it beneficial to hide Like counts on Instagram, as well as perhaps why there’s still such a demand for text-first platforms like Threads and X. Even if social media isn’t single-handedly responsible for torpedoing international morale, there seems to be little doubt it’s a double-edged sword for many users.

Zoom has made a U-turn on two key company policies in the space of a few days, with the video conferencing provider making decisions that impact both staff and customers.

Most recently, the company was forced to clarify whether user data from Zoom calls and chats would be used to train its AI tools, and even changed its Terms of Service to reflect that.

Just days before that, despite being the proverbial poster child of the remote working revolution, Zoom decided that its employees need to start coming into their offices in person. The company now believes that a hybrid approach to working is more effective.

Zoom Updates its Terms of Service After Backlash

Zoom confirmed this week that it will not use customer audio, video and chat data to improve its artificial intelligence products, tools, and systems.

Alarm bells were sounded after StackDiary reported over the weekend that changes made to the company’s services in March 2023, gave the company far-reaching permissions to use user data in machine learning processes. This led to a clarificatory change to the company’s terms of service.

The privacy policy says that Zoom has “…perpetual, worldwide, non-exclusive, royalty-free, sublicensable, and transferable license and all other rights required or necessary to redistribute, publish, import, access, use, store, transmit, review, disclose, preserve, extract, modify, reproduce, share, use, display, copy, distribute, translate, transcribe, create derivative works, and process Customer Content…”

It also says that, by downloading a Zoom product, “you consent to Zoom’s access, use, collection, creation, modification, distribution, processing, sharing, maintenance, and storage of Service Generated Data for any purpose… including for the purpose of… machine learning or artificial intelligence (including for the purposes of training and tuning of algorithms and models), training, testing, improvement of the Services, Software, or Zoom’s other products…”

In response to the uproar, Zoom published a blog post saying it wouldn’t use customer audio and video data, and added this to its terms of service:

“Notwithstanding the above, Zoom will not use audio, video, or chat Customer Content to train our artificial intelligence models without your consent.”

Zoom Calls Staff Back To Office

Just a few days prior, Zoom reversed another decision, asking employees working within fifty miles of one of their offices to shift to hybrid working arrangements and come into the office at least two days a week.

The new policy will come into force during August and September and is thought to affect around 7,400 employees, CBS reports.

Zoom has said that their video conferencing technology will remain integral to the way their employees work but admitted in-person communication is an effective way to work. They join other tech companies, like Google, in bringing employees back into workspaces.

While some have pointed out the irony of Zoom’s desire to get its employees back into the office, others have dubbed it as a real sign that enthusiasm for remote work is dwindling as more and more companies end remote work in the US and all over the world.

Cybersecurity researchers have shown that an AI model can work out what people are typing based on the sounds that different keys make when they’re pressed, and could feasibly be used to guess victims’ passwords during video conferencing calls.

Considering the number of apps the average user hands over microphone access to, as well as the extensive use of software like Zoom by remote employees, these findings are extremely concerning. Passwords and other sensitive information could easily be uncovered using this sort of technique.

A deep learning model with the power to identify keystrokes in this way is just one of the latest instances of AI scams, fraud, and cyber attacks being linked to the booming new technology.

AI Can Recognize Almost Every Keyboard Key You Press

In a recently-published study, UK-based researchers Joshua Harrison, Ehsan Toreini and Maryam Mehrnezhad found that a deep learning model could classify 95% of keystrokes made on a 2021 MacBook Pro recorded on a nearby iPhone 17.

They trained the model by pressing 36 keys a total of 25 times each, in order to help it recognize the individual waveforms produced by individual keystrokes.

Along with the impressive – and worrying – results achieved via the iPhone audio recording, “when trained on keystrokes recorded using the video-conferencing software Zoom, an accuracy of 93% was achieved, a new best for the medium”, the researchers said.

Interestingly, on the few occasions the deep learning model used by researchers was mistaken in its classification, it was only a few keys away from the right answer. This suggests the position of the keys on the keyboard contributes significantly to unique audio profiles.

The researchers say that their results “prove the practicality of these side-channel attacks via off-the-shelf equipment and algorithms.”

What Are Side-Channel Attacks?

Side-channel attacks are a category of cyber attacks that take advantage of extra information available via channels and mediums created by the physical implementation of a system and its hardware.

The capturing of audio in a video call is just one example of a side-channel attack. Van Eck Phreaking – a different kind of side-channel attack – involves utilizing equipment that can pick up electromagnetic emissions emitted by LCD displays to extract data about what’s on the screen.

How to Protect Yourself From The Threat

Some recommendations of what you can do to combat cyber attacks facilitated by AI tools with the above capabilities are put forward by the trio of researchers.

Perhaps the easiest to implement action is simply using multiple cases in your password. The AI model was good at recognizing most keystrokes but struggled with the shift key, according to Ars Technica.

Other recommendations include changing the rhythm at which you type and adding decoy keystrokes when typing while on video calls. Using a password manager to help juggle multiple combinations also comes highly recommended, as it’s an effective way to limit the damage if one of your accounts is compromised.