Another day, another artificial intelligence feature that can help businesses sell online: Amazon is announcing an agentic AI update to its Seller Assistant platform.

AI continues to evolve at break-neck speeds, with every business under the sun making an effort to roll out the innovative technology on their platforms. From generative tools for developing copy to chatbots that can guide you through setup, you can find them pretty much everywhere in 2025.

Amazon is, of course, among those making a big push to make the technology part of its platform, and a big change just came to one of its more valuable assets.

Amazon Adds Agentic AI to Seller Assistant

Announced in a company blog post, Amazon is adding some seriously helpful functionality to an AI offering called Seller Assistant. If you’re unfamiliar, Seller Assistant is a generative AI tool announced by Amazon last year that provided users with quick answers to questions, directing users to helpful resources to solve their problems.

With this recent announcement, though, Seller Assistant is getting a big upgrade. The agentic AI improvement will give the platform a lot more power, allowing it “to reason, plan, and help take action with a seller’s permission.”

Seller Assistant is available at no cost to all sellers in the US right now, but it isn’t available outside the country just yet. Amazon does promise that it “will be rolled out to other countries in the coming months.”

What Can Seller Assistant Do Now?

The initial release of Seller Assistant last year wasn’t too exciting, with simple generative tools that can be found pretty much anywhere. Now, though, Seller Assistant is getting an upgrade that will provide some serious firepower for businesses looking to sell online.

Here are some of the things that Seller Assistant can now do?

• Actively monitor inventory levels
• Deliver a analysis of issues requiring immediate attention
• Automate compliance navigation
• Develop professional ads through conversational prompts
• Suggest promising new product categories
• Recommend advertising and marketing strategies
• Develop comprehensive growth plans

Will Seller Assistant Actually Help?

Selling online can be pretty hard in 2025.

With the tariff situation causing undue strife and the economy forcing many to tighten their purse strings, sellers need every tool they can find to make a difference. Luckily, it sounds like the improvements to Seller Assistant could do a lot to actually improve your revenue.

“I’ve been using Seller Assistant almost every day now, and it has become my own personal business consultant. From FBA to sales reports, it knows how to support my business in Amazon’s store and provides me with the answers and actions that I need immediately, rather than me needing to dig around and spend time finding what I need. I had no idea it could do all of this; it even understands sales velocity.” – Alfred Mai, founder and CEO of ASM Games in the Amazon press release

Let’s be honest, AI hasn’t fully proven its value to businesses just yet. Many are reporting that the technology provides almost no return-on-investment so far, leading some companies to dial back AI adoption.

Still, if Seller Assistant can actually deliver this kind of assistance to businesses trying to stay afloat in 2025, it could be poised for a comeback for those trying to sell online.

Texas is planning to take its enforcement of English language proficiency for truckers to the max, with the governor announcing that the zero-tolerance policy will be enforced across the state.

The trucker shortage has become a big issue across the logistics industry. A lack of labor options has many businesses struggling to keep up with demand, leading to delays that impact businesses as well as consumers.

That doesn’t seem to bother the governor of Texas, though, who plans to take even more truckers off the road with strict enforcement of English language proficiency rules that could negatively impact the workforce.

Texas Governor Announces Strict Enforcement for English Proficiency Requirements

Announced in a statement from the governor’s office, Greg Abbott has stated that the Texas Department of Public Safety (DPS) plans “to strictly enforce the English language proficiency requirements” on Texas roadways moving forward.

“Truckers play an instrumental role in Texas’ robust economy and in keeping our highways safe. Every commercial driver license operator on Texas roadways must be able to communicate clearly in English to ensure compliance with traffic laws, follow safety directions, and prevent accidents.” – Texas Governor Greg Abbotts

It is unclear what exactly these strict enforcements will look like, but the statement from the Governor noted that troopers “will conduct English Language Proficiency reviews for all commercial license operators on Texas roadways,” which could mean that roadside stops of potential non-English speakers are on the table.

Why Is Texas Cracking Down on Non-English Speakers?

There are understandably some safety concerns about non-English speakers hauling freight on US roadways, but the problem seems to have gotten a lot of attention lately, and you can probably guess why.

That’s right, President Trump has used the issue to further his anti-immigrant agenda, signing an executive order on June 25th that puts strict English language proficiency requirements on truckers across the country.

The executive order has been enforced in Texas quite a bit, with “approximately 445 commercial vehicle drivers in Texas” being sidelined on the road for English proficiency violations.

The Impact of English Proficiency Requirements on the Logistics Industry

As you can imagine, these English language proficiency requirements are putting a bit of a strain on the logistics industry. In fact, our Pulse data found that more and more professionals see “adapting to regulations” as a primary concern for their businesses.

On top of that, the national trucker shortage is just as prevalent in Texas, which means that these truckers coming off the road will likely contribute to the worsening workforce numbers.

Suffice to say, Texas may run into some serious logistical problems moving forward if they can’t figure out a way to add drivers to the state, rather than removing them because they can’t speak English.

Layoffs are expected over at Fiverr, with 250 workers expected to lose their jobs as the company bids to go “AI First.”

Fiverr is not the first company to align itself more substantially with artificial intelligence, with Duolingo doing the same after it announced it would be laying off contract workers in the process.

However, it may be the case that companies are going AI-centric too soon, especially as there are still a lot of questions about whether the technology is equipped to take on and replace human workers, without harming operations or overall profit.

250 Workers to Lose Jobs As Fiverr Looks Towards AI

Online marketplace Fiverr has announced it will be laying off 250 workers, in a shift that will allow it to focus more on AI.

As of December of last year, the company had 762 employees, so the layoffs will account for around 30% of its workforce. The company hopes it will be able to reinvest part of the savings from the layoffs into the business.

In a letter to employees that CEO Micha Kaufman shared on LinkedIn, he stated: “We are launching a transformation for Fiverr, to turn Fiverr into an AI-first company that’s leaner, faster, with a modern AI-focused tech infrastructure, a smaller team, each with substantially greater productivity, and far fewer management layers.”

‘AI First’ Is the Modern Company Epithet

Fiverr is not the first company to boldly proclaim itself to be “AI First,” a structure that sees businesses place AI at the core of its operations. Not long ago, Duolingo CEO Luis von Ahn came under fire after he claimed the language-learning platform would be replacing some of its contract workers with AI, in a quest to also go AI First.

Although, von Ahn would later restyle his original statement, to claim that the company doesn’t plan to replace full-time employees with AI.

Other companies have made similar pledges. Shopify operates an ‘AI First’ hiring rule, that has managers prove that AI can’t do a job before they are allowed to hire for it. This is further evidence that companies are not only using AI tools to improve productivity, but are wanting to see these tools at the center of their operations.

Too Much AI Too Soon?

Other than in cases where businesses have implemented AI or replaced workers with it, and then hastily changed their minds, there is still some mystery around how an AI First company will operate. In a report earlier this year, Microsoft predicted that future businesses that want to survive the AI revolution should look to become a “Frontier Firm,” which sees humans managing teams of AI agents.

However, while Microsoft urged businesses to act quickly, is this too quick? Especially since there are still mixed studies around AI productivity, and plenty of examples where AI has been deployed only to generate unpredictable results.

Similarly, there is the possibility that companies are simply jumping on the bandwagon of AI adoption, rather than finding out how it can help their business specifically. Studies have shown that peer pressure is playing a role in AI spending. This could be further prove that yes, the technology is moving impressively fast, but is it too fast to end up being worthwhile?

Zoom CEO Eric Yuan has joined Bill Gates and Jensen Huang in suggesting AI deployment in companies could lead to a three or four-day workweek, because of the potential time it could free up for employees.

Yuan has also predicted that while some jobs will end up disappearing, new jobs will appear in their place as the technology develops.

While the four-day workweek hasn’t been entirely ruled out by senior leaders, it has been overshadowed by frequent return-to-office mandates. However, automation and AI could be the push the initiative needs to become more widely accepted.

Zoom CEO Predicts Three or Four-Day Workweek Thanks to AI

The CEO of Zoom, Eric Yuan, has predicted that AI chatbots and agents could enable companies to introduce a three or four-day workweek. Yuan believes that if AI is able to make our lives easier, then there’s a big chance that workers will be able to clock in for fewer days a week.

He told The New York Times in a recent interview that, “Every company will support three days, four days a week. I think this ultimately frees up everyone’s time.”

While the four-day workweek has been trialled in some European countries, such as Iceland and Belgium, it’s yet to become a mainstream reality for US businesses. However, developments in AI could be what the initiative needs to charge forward.

Experts Agree AI Will Enable Fewer Working Days

Yuan is not the only tech leader who has expressed ideas about a shorter working week as a result of AI. In an interview on The Tonight Show with Jimmy Fallon, Bill Gates said that he didn’t think we’d need humans for “most things” in the next ten years, freeing up humans to work less than five days.

Even Jamie Dimon, CEO of JPMorgan Chase, a company within an industry that likes to put its workers through long working hours, has expressed interest in the idea. He told Bloomberg TV in 2023 that AI will lead to people working “3 and a half days a week”.

However, it may not be all positive news for workers. In fact, Yuan explained that some job opportunities will be gone, especially for entry-level candidates, as AI continues to develop. But he also expressed some hopes that new jobs will pop up in their place. Engineers, for instance, could get jobs managing AI code or AI agents.

Could AI Push The Four-Day Workweek Into Fruition?

Tech.co’s 2025 Impact of Technology on the Workplace report found that the idea of a four-day working week has picked up, with 38% of senior leaders being open-minded about implementing the initiative, compared to 23% last year. With AI now making its way into workflows, this could be the fuel the initiative needs to become more mainstream.

Of course, this will depend on how much AI develops over the next few years. It’s still not entirely certain how productive AI can be. Some studies have found that AI can increase productivity, whereas others have questioned whether it only makes users feel they are being more productive.

Not only will AI have to prove itself, but business leaders will have to learn to put more trust into the technology, particularly if it’s going to take on a significant amount of tasks, or even supplement a full day of work for an employee. While things certainly seem to be moving at a fast pace, it may be a while before we see a level of AI use within companies to the extent where employees can be spared for a day or two.

FedEx will raise the cost of shipping packages in January 2026, the delivery and transportion service has announced.

The exact rate increases will vary, but the standard list rates for US package shipping will go up by 5.9%, in addition to a handful of other surcharges.

Shipping costs have been earning headlines across 2025, due in large part to the US import tariff increases that have been issued, paused, and re-issued since February. However, these prices are identical to past annual increases from FedEx: They’re not tariff related.

The FedEx Price Increases

The 5.9% increase for standard rates will impact US exports and imports alike.

Additional surcharge increases are on the way in January 2026, too. Here’s the breakdown for those, via Supply Chain Dive:

• Additional handling for US package services is now $43.50 (up from $46.00)
• Delivery area US package services (Residential) is now $6.20 (up from $6.60)
• Delivery area US package services (Commercial) is now $4.20 (up from $4.45)
• Oversize (Zones 3-4) FedEx Home Delivery is now $260 (up from $275)
• Residential delivery US package services is now $6.55 (up from $6.95)
• US inbound processing Fedex International Ground $2.50 is now (up from $2.65)

FedEx has the documentation for its official 2026 standard list rates available online.

Additional rates for other services including Ground Economy and Ground Multiweight are also going up, although FedEx has not yet specified what those increases will be.

It’s Not Impacted by Tariffs

FedEx increased standard rates by 5.9% in 2024 and in 2025. In other words, this is the third year in a row in which the company has steadily increased costs by the exact same amount.

Tariffs may impact shipping costs down the road, but this particular increase is just a cost of going business. In fact, delivery competitor UPS also boosted its rates by 5.9% last year.

FedEx’s International Processing Fee Rose Recently

That’s not to say that the many regulatory shakeups from the current government administration haven’t impacted FedEx at all: Last month, the company did increase its international processing fee in a response to the federal government’s removal of the de minimis exemption, which had previously exempted imported packages under $800 from any duties.

To be specific, the company boosted its international processing fee from $1.50 to $2.50 on August 18th.

That fee hike applies to all packages shipped through International First, International Priority Express, International Priority, International Economy, International Express Freight, and International Ground services.

The global shipping giant previously laid off more than 600 employees in Memphis.

ChatGPT launched in late 2022. By the middle of 2025, the platform was fielding 18 billion messages per week from 700 million users. Needless to say, that’s a big change.

Now, a new study has dug into the data on exactly how that generative AI chatbot is being used. Most conversations with ChatGPT are for personal reasons, not work-related ones, it turns out.

Another insight from the study — which analyzed over 1.5 million chats — revealed that one out of every four conversations centered around information gathering. Here’s what else to know.

More and More Personal Reasons to Use ChatGPT

The giant new 62-page research paper, done in collaboration with Duke and Harvard universities and which is available online, found that 73% of all conversations during June 2025 were for non-work reasons, in contrast to just 27% being for work reasons during that time period.

That’s a shift towards personal use when compared to how people were using ChatGPT exactly one year earlier: In June 2024, 53% of uses were non-work-related, while 47% were for work reasons.

Perhaps better work performance was the reason why many users first checked out ChatGPT, with its benefits later slipping into their personal lives as well.

That’s just one interpretation of the data, however.

10% of Adults Globally Are Using ChatGPT

It’s also possible that the last year saw a demographic shift in the types of people who began using ChatGPT. After all, the tool has a big pool of users: About 10% of the total global adult population is interacting with ChatGPT, the study found.

The last few years have seen a gendered shift in use, too: In early 2023, about 80% of users had “typically masculine” first names, but in June 2025, 52% of users had feminine names. That’s a gender split that much more closely maps to the actual population of the planet.

It’s clear that personal use of AI is higher than ever, and that more people than ever are trying it out.

What Are People Using ChatGPT for?

The three most common uses of ChatGPT were, in order: “Practical guidance,” “Writing,” and “Seeking information.”

“Practical guidance” was the number one use case, representing 28.3% of all chats. That’s a category that’s wide enough to encompass a huge range of how-to advice, from schoolwork aid to workout tips.

However, not that many people are turning to the bot for relationship advice, even if AI girlfriends are making headlines: Just 1.9% of the conversations are about relationships, feelings, and “discussing personal reflections.”

What’s the ultimate impact of this widespread ChatGPT consultation? We’ll no doubt see many more studies on the subject in the years come, although we already have a few to consider, from a recent MIT study that found AI tools are reducing “engagement in deep, reflective thinking,” to a research paper from Microsoft and Carnegie Mellon University that raised “concerns about long-term reliance.”

Generative AI tools continue to be used by more and more businesses around the globe, new data confirms. In fact, a massive 77% of enterprise customers’ AI use is for fully automating tasks.

AI company Anthropic has just released a large-scale study on AI use, detailing exactly which countries are using its Claude AI tool the most, as well as which types of tasks are most likely to prompt them to turn to the emerging technology.

There are plenty of interesting tidbits buried in the full study. Here’s our look at the best.

AI Use Has Doubled Over the Last 2 Years

Anthropic’s new study is an analysis of how often its Claude AI tool has been used across over 150 countries and all 50 US states, as well as what it has been used for.

Generative AI has been massively hyped in the tech world for years now. But it takes a lot longer for a new technology to actually trickle out to widespread use. As the new study puts it, AI automation is continuing to pick up speed.

In the US, the report found, 40% of employees report using AI at work in 2025, up from 20% in 2023 two years ago.

“AI is being used to fully (or ‘directively’) automate more and more tasks over time. In just nine months, we saw directive automation jump from 27% to 39% of all conversations. For enterprise customers, that figure is 77%.” – Anthropic

Within the US, DC Uses AI the Most

Thinking Silicon Valley would put California at the top of the list of the most AI-centric US states? Think again.

Washington, DC has the highest per-capita AI usage in America, with an amount of Claude AI users that’s nearly four times higher than you might expect based on its population size.

Granted, California clocks in as the third most frequent AI using state, after Utah took second place. Here are the top ten states, in order:

1. Washington, D.C.
2. Utah
3. California
4. New York
5. Virginia
6. Washington
7. Massachusetts
8. Colorado
9. Vermont
10. Nevada

The study further noted that each state’s AI use centered on business practices that were more common within that area.

“Some states’ usage jumps out as characteristic of the states themselves: in Colorado, use focuses disproportionately on travel and event planning, and in DC, people are especially likely to use Claude for document editing and career advice.” – Anthropic

Higher Incomes Drive More AI Use

Zooming out to get a global look at AI adaptation reveals additional general insights. First, increasing adaption strongly correlates with higher incomes: Singapore and Canada have strong economies and above-average Claude usage, while emerging economies, including Indonesia and Nigeria, use AI less.

In addition, the study concluded that high-adaptation countries tend to turn to AI for a more diverse range of tasks.

According to the study, “Lower-adoption countries tend to see more coding usage, while high-adoption regions show diverse applications across education, science, and business.”

Will AI adaption rates continue rising as rapidly as they have over the past two years? We’ll have to wait for Anthropic’s next study to learn the answer.

They said AI would make life easier, and it is… for scammers, of course. A new tool on the dark web — dubbed SpamGPT — is enabling cyber criminals to enact massive phishing scams that are, unfortunately, very effective.

AI has experienced its fair share of bad press over the last few years. From egregious errors to full-on hallucinations, the technology isn’t exactly at the top of its game just yet.

This is even worse, though, as AI technology is clearly being used to make scams even harder to spot in 2025.

What Is SpamGPT?

SpamGPT is a cybercrime toolkit that is being used to generate and dispense massive phishing campaigns to users in 2025.

The AI tool has been discovered on underground forums like the dark web, sold as a “spam-as-a-service” platform that cyber criminals can use to develop more sophisticated attacks on unsuspecting individuals.

SpamGPT is reportedly being sold on these underground forums for $5,000.

How Does SpamGPT Work?

SpamGPT works like any email marketing platform. It offers tools like SMTP/IMAP email server, email testing, and campaign performance monitoring in real time. There’s also an AI marketing assistant named KaliGPT, which is built directly into the dashboard for assistance.

However, where it differs from email marketing platforms is that it is specifically designed for creating spam and phishing emails to steal information and financial data from users.

More specifically, SpamGPT is “designed to compromise email servers, bypass spam filters, and orchestrate mass phishing campaigns with unprecedented ease,” according to a report from Varonis, a data security platform.

How to Protect Yourself From SpamGPT

While SpamGPT sounds pretty terrifying, the reality is that you don’t have much to worry about as a business if you’re already taking phishing scams seriously.

SpamGPT mostly just increases the number of emails cyber criminals can send out, while also making the phishing scams a bit more convincing. As a result, any business that has trained its employees and put in safeguards against phishing should still be safe.

Unfortunately, that’s not necessarily the case. In fact, a recent study found that the majority of businesses could be vulnerable to these attacks, with 98% of owners not being able to recognize all the signs of a phishing attack.

UPS is helping out the little guy this holiday season: It’s forging a new partnership with American Express to provide small and medium-sized businesses with discounted shipping rates through the beginning of next year.

It’s no secret that many logistics businesses are hurting right now. Between the persistent trucker shortage and the complicated tariff situation, many companies are struggling to stay afloat to see how it all shakes out. Even UPS itself has seen significant layoffs over the last few months.

That could be why they’re taking a big swing to attract small and medium-sized businesses to their operation, with some hearty discounts that could actually make a difference for many companies in 2025.

UPS & American Express Partner for Lower Rates

This week, UPS announced a partnership with American Express that aims to provide small businesses with a bit more wiggle room when it comes to shipping their products.

More specifically, merchants that use the Business Savings Suite from American Express will now officially get some exclusive discounts on air, ground, and international. On top of that, the rates get even lower the more you ship, which could help small businesses that are trying to grow.

The lowered rates come at the perfect time for businesses that are likely enduring the stress of tariffs and preparing for a stressful holiday season. The rates will be available throughout the busy season too, as they won’t expire until February 28th, 2026.

Why Is UPS Offering Lower Rates to Small Businesses?

It’s not easy for small businesses right now, particularly those that rely on shipping products. The tariff situation alone is causing some businesses to shutter their windows, and UPS seems to want to help alleviate that as much as possible.

“While our customers who have scale may be able to thwart the impact of rising costs due to tariffs, many of our SMB customers may not.” – Carol Tomé, CEO of UPS, on a July earnings call

On top of that, attracting small business owners is very much in the interest of UPS, as well. Many are looking for alternative shipping options after Trump got rid of the de minimis exemption and UPS could really cash in if they represent an attractive offer.

Tariff Volatility in the Logistics Industry

While this partnership between UPS and American Express is good for small businesses, the reality of the tariff situation for logistics businesses is far more complicated.

In fact, UPS recently announced new international processing fees for packages shipped into the US, pointing to a lot of volatility in an industry doing its best to keep up with the constantly shifting regulations surrounding the shipping world in 2025.

Logistics businesses can expect more volatility in the coming months and even years, as these policies start to take effect and really impact the economy.

Is the AI honeymoon over? Data from the US Census Bureau says it might be, with AI adoption among larger companies reportedly on the decline for the first time since 2023.

For the last few years, AI is virtually the only thing you’ve heard about from the tech world. Models, chatbots, widgets, and extensions have been added to every software under the sun to “improve productivity” beyond our wildest collective imagination.

Now, though, it seems like many companies are dialing back, likely due to the AI return-on-investment being a lot more lackluster than promised.

AI Adoption Is Finally Slowing Down

According to data from the US Census Bureau, fewer large companies are adopting AI into their operations. This represents the first significant and consistent decline in AI adoptions rates since the technology become massive popular in 2023.

More specifically, the data points to businesses with over 250 employees adopting AI significantly less than just a few months ago, with businesses with between 100 and 249 employees and between 50 and 99 employees both seeing a more modest decline.

This data comes from a biweekly survey conducted by the US Census Bureau and collected by Apollo Academy. The survey covers professionals from 1.2 million firms in the US, and specifically asks “whether a business has used AI tools such as machine learning, natural language processing, virtual agents or voice recognition to help produce goods or services in the past two weeks.”

Why Are Fewer Businesses Adopting AI?

Businesses have been pushing an investment in AI like crazy over the few years, so this could just be a natural correction to the overexcitement about the technology. Or, it could be something worse.

According to a recent study from MIT, the majority of businesses are not being inundated by revenue as a result of AI, with 95% of businesses noting that they have seen no return-on-investment as a result of the technology.

Simply put, if technology isn’t making money for a business, they aren’t going to keep using it. And in this case, there are a lot of larger businesses that aren’t seeing their investment in AI lead to anything.

How to Use AI to Improve Your Business

Just because large businesses aren’t taking advantage of AI as much as they used to, doesn’t mean you shouldn’t be. There are some real use cases for the tech that can bolster the chances of a small business to succeed.

AI website builders are some of our favorites at Tech.co. They can create a fully functionality website for you in just a few seconds with just a few questions. Plus, they’re often equipped with helpful AI generators for copy and images, so you can really get it to look how you want.

Suffice to say, there are plenty of ways that AI can be used to improve your business. It’s just a matter of finding out what they are and getting on board.

Microsoft is the latest tech giant to throw its multi-trillion-dollar market cap behind a return to the office: The tech company will be requiring employees to show up on-site at least three days per week.

The new policy will start rolling out for employees at Microsoft’s headquarters in Washington State on February 2026.

It’s an apparent end to Microsoft’s reputation as one of the most remote-friendly tech giants, and a yet another sign that the tech industry is closing the door to remote workplace accommodations that proved so helpful when the 2020 pandemic began.

What to Know About the Microsoft RTO

The first employees to be forced back to the office are those living within 50 miles of company headquarters in Redmond, Washington in February of next year. More of Microsoft’s US and global locations will follow.

Those in the first round of RTO-mandated employees will have until Friday, September 19 to request exemptions from their managers.

It’s a noteworthy shift: As GeekWire notes, under the company’s current hybrid work policy, most employees can work remotely for up to half the time without approval.

Writing in a memo on the company blog, Executive Vice President and Chief People Officer Amy Coleman states that “this update is not about reducing headcount.”

Microsoft’s New Perspective on Remote Work

Coleman’s memo on the policy change cites one common argument on the value of in-person work: That it promotes collaboration. As Coleman puts it, “when people work together in person more often, they thrive — they are more energized, empowered, and they deliver stronger results.”

The counterargument — one particularly relevant to those who are parents to young children, those caring for elderly relatives, or those with disabilities — is that any workplace accomodations that reduce employee stress are what truly promotes more energized and empowered workers.

In an article discussing the new RTO mandate, Business Insider notes that Microsoft has revised its position on the topic of remote work, writing that “while working on the new RTO policy, Microsoft appears to have scrubbed a blog post that once heralded the benefits of remote work for retaining employees and boosting their productivity.”

Microsoft’s Dwindling Interest in Supporting Fully Remote Work

Here at Tech.co, we’ve kept an eye on the number of fully remote positions open at Microsoft over the past few years, and a trend has become clear.

In August 2024, 883 fully remote jobs were available. By November, this had fallen to 400-something postions, a trend that continued in December, when we spotted 417 remote positions open. In February 2025, just 313 work-from-home jobs were open.

Today? That number stands at just 89 positions.

In 2026, it looks set to drop further.

Fully remote positions will likely never go away entirely. Even at major companies with RTO policies, exceptions will be made for certain employees with the right leverage. However, Microsoft is finally joining the majority of tech companies that are pushing their employees towards in-office work.

Reddit, Yahoo, Medium, wikiHow, and many more content-publishing websites have banded together to keep AI companies from scraping their content without compensation.

They’re creating “Really Simple Licensing” (RSL), an open licensing standard that allows online publishers to set up machine-readable licensing terms for their content.

Huge AI companies have been getting in legal trouble for allegedly stealing massive amounts of data off the internet for years now. This new infrastructure hopes to give them a simple way to pay the people who are providing the future training data that AI companies everywhere will need to stay competitive.

What Is Really Simple Licensing?

The RSL Standard is an upgrade on the robots.txt protocol, which is what allows websites to determine which areas will or won’t accept the content-scraping bots that AI companies use to collect data, which they then feed to their AI models to create generative AI chatbots like ChatGPT or Gemini.

With RSL, sites can embed licensing and royalty requirements directly in robots.txt files. They can do the same with books, videos, or other training datasets, too.

With the RSL Standard, online publishers can ask AI companies to pay an ongoing subscription or a one-time pay-per-crawl fee.

Currently, a few specific publications (like The Wall Street Journal’s owner or The New York Times) have set up licensing agreements with certain AI companies, as The Verge notes in its coverage. This new standard could let many other publishers streamline and speed up that type of process.

Content Publishers Hoping to Fight Back

The list of publishers who have thrown their weight behind this new standard are a murder’s row of the websites that stand to lose the most from the birth of AI-generated content.

Reddit, Yahoo, Medium, Quora, People Inc.: They’re all websites that many people turn to for answers to everything from niche hobbies to science homework to do-it-yourself home repairs.

Today’s chatbots need up-to-date versions of that type of data in order to deliver accurate responses — even if this sometimes leads to hallucinations too, as in the case of a Google AI overview that turned a sarcastic reddit post into a serious recommendation to put glue on a pizza.

Will This Stop AI Companies From Getting Sued Over and Over?

Back in 2023, OpenAI  dealt with one $3 billion class action lawsuit related to the “secret scraping” of data used to train ChatGPT models, in addition to a separate suit against both OpenAI and Meta, from authors accusing them of copyright infringement.

Today, similar cases are still making headlines. Anthropic just settled a class action lawsuit over hundreds of thousands of pirated books that were used to train AI chatbots. The $1.5 billion payout may still be subject to changes.

However, it’s easy to see why AI companies and online publishers are both interested in coming to a compromise.

If everyone wants to play ball, this new standard may be the best chance yet that AI-skeptical writers have of keeping their content from being scooped up by AI.

Anthropic’s Claude is the latest AI tool to roll out new functions: The chatbot will soon allow users to create and edit Excel spreadsheets, documents, PowerPoint slides, and PDFs.

Users will be able to create these new materials entirely with written prompts, sent through the Claude chat interface.

This new AI feature announcement comes soon after ChatGPT’s latest free feature debuted, allowing users to group their prompt conversations into folders, and right on the heels of Anthropic’s 1.5 billion settlement over a class action lawsuit from book authors upset over unauthorized data training.

Can You Access the New Features Yet?

The file creation feature is rolling out over the next few weeks to Pro users. It’s already available to Max, Team, and Enterprise plan users.

It’s an opt-in function currently: To find it and turn it on, you’ll need to head to Settings, look for the experimental section under Features, and look for “Upgraded file creation and analysis.”

According to ZDNet, Anthropic is positioning the tool as a way to avoid the chores that go into formatting and reformatting media like spreadsheets and slides:

“Some examples Anthropic provided included turning raw data into cleaner projects with analysis, charts, and written insights; building spreadsheets without having to worry about formulas; and transforming your content from one format to the other, for example, a PDF into Slides.”

Will These Tools Change Your Workflow?

The new tools will allow users to change up file formats, turning a doc file into a PDF with a single command. (Granted, that’s already possible with a dropdown menu button in Google Docs, too.)

Users might use the tool to reformat their school homework. However, the same tool can fully generate the writing prior to reformatting it, so it’s possible that this is just one more way for students to easily create an essay or other homework rather than writing it themselves. This just might be yet another AI nightmare for school teachers everywhere.

One word of warning from Anthropic’s press release: Claude requires internet access to create or edit your personal files, and this “may put your data at risk.”

AI Technology Continues Growth

The new functions from Claude aren’t the only evidence that AI tools are continuing their highly-hyped rapid growth.

Huge AI companies continue to cut new deals to further expand. Most recently, Oracle and OpenAI inked a truly huge $300 billion cloud computing deal that will sink 4.5 gigawatts of energy into AI over the next five years.

Plenty of concerns are still remaining on AI’s rocky road to long-term sustainability, and many experts are predicting a dot-com style bubble will pop before the technology truly settles into use. Until then, we’ll keep debuting new features while investing ever more energy and money into the tech.

Accounting firm PwC has admitted it will be lowering its entry-level recruitment this year in its UK branch, taking on 200 fewer graduates. PwC recently announced similar news across its US operations, too.

Marco Amitrano, the company’s UK chief, has said the company’s lowering numbers reflect a slowdown in investment, hiring, and deal-making across the country. However, he has also nodded to AI as a factor in reshaping current roles.

PwC will therefore join the ranks of companies looking towards AI, although what this looks like down the road for graduates and entry-level candidates, nobody seems completely sure.

PwC Reduces Entry-Level Hiring, Admits Graduate Roles “Under Pressure”

Big Four accounting firm PwC has confirmed it will be lowering entry-level recruitment this year to its UK branch, taking on 1,300 entrants to its graduate programs, compared to the 1,500 entrants it took on the year before.

While Marco Amitrano, PwC’s UK chief, has said that the company will “always be a large employer and training ground for young people,” its hiring habits have changed, based on the impact of AI and a sluggish economy.

Back in August, Business Insider reported that graduate hiring would also be slowing in PwC’s US branches, adding more uncertainty for entry-level candidates. The company plans to cut graduate hiring by a third over the next three years.

AI Is Reshaping Roles, But This May Not Be Forever

Amitrano, in an op-ed for The Times, said that AI is definitely reshaping roles, with the development of new tools and the investment in skills “offsetting more serious disruptions” to the market, but this disruption may not last forever.

“Our research shows that job postings for AI-exposed occupations are growing at a slower pace compared to those with lower exposure – and this gap is widening.” – Marco Amitrano, head of PwC’s UK practice.

However, Amitrano is optimistic about the future of AI. He wrote, “Most technologies, if not all, have boosted both productivity and wages. Indeed, the integration of AI with other technologies, products and services has the potential to create whole new industries. But potential doesn’t convert to progress on its own; it requires sustained and serious investment in tech, infrastructure and training — underpinned by progressive public policy.”

PwC Joins Amazon and Coinbase in AI Future

PwC is not the first business to look towards AI for the future of the working world, and to suggest that this will affect hiring patterns. Amazon CEO Andy Jessy said back in June that AI will lead to there being fewer workers at the company.

Some companies have taken on a more hardline approach. Coinbase recently admitted to firing engineers who failed to sign up for the company’s AI tools after a week of being asked to, a less than subtle way of saying it’s the AI way, or no way.

However, how much of this is in the best interest of these companies? A study from MIT recently found that 95% of generative AI enterprise pilots at companies fail. At the moment, Amitrano writes that he is doing what other businesses are doing, “watching and waiting.”

A recent study from cloud software company Fastly has found that senior developers are much more likely to use AI-generated code compared with junior developers, as well as deploy this code for use in production.

Senior developers also feel that using AI tools makes them work faster, more than junior developers do. However, senior engineers were more likely to spend time correcting AI code compared with their less-experienced counterparts.

AI has without a doubt changed the landscape for developers, shaking up the job market and altering the day-to-day tasks of the role. While evidence is mixed on whether AI improves efficiency, this study certainly suggests that developers want to keep AI around for the time being.

Study Finds Senior Developers Use More AI-Generated Code Than Juniors

A new study from Fastly has found that senior developers are using more AI-generated code than junior developers, and are deploying more of this code to be used in production.

Overall, nearly one third of the senior developers polled revealed that over half of the code they ship is AI-generated. This figure is more than double the rate of junior developers, with only 13% reporting the same thing.

Findings point to a higher rate of AI use among senior developers, and suggests there may be a higher level of trust in AI code for these experienced developers. Fastly’s study consisted of 791 professional developers, made up of senior developers (those with 10+ years of experience) and junior developers (those with less than two years’ experience).

Senior Developers Say AI Makes Them Faster, But Spend More Time Correcting Mistakes

AI is certainly good at taking over autonomous and repetitive tasks. However, the sectors and areas where it can save employees the most time is still not entirely clear. In the study, senior developers were more likely to view AI as a net time-saver.

In fact, senior developers had a stronger enthusiasm for AI as a tool to help them work faster, with 59% saying AI sped up their working process. Comparatively, only 49% of junior developers said the same.

However, these findings come hand-in-hand with the fact that senior developers were more likely to spend time correcting AI-generated code. This highlights a gap between how fast a developer believes they are working, and the reality of having to make changes.

Indeed, 28% of developers said they spent so much time editing AI code that any potential time-saving benefits were made void. Only 14% said they rarely needed to make big changes to work generated by AI.

Does AI Make Developers Faster and More Efficient?

Senior developers, the study points out, have plenty of experience to lean on when dealing with AI-generated code, compared with juniors. What may be the case is that senior developers are more capable of finding vulnerabilities or mistakes, and believe making these changes is more efficient than writing the code themselves from scratch.

However, it still may not be the case that tools such as Copilot or Claude make software engineers faster and more efficient. In another trial from Fastly, open-source developers took 19% longer to complete tasks when using code assistants. This suggests that what may feel fast to a developer — maybe because they are getting all of the work done, pretty much immediately — doesn’t equate to tangible productivity gains.

Another notable element of this study is its clear stance on job satisfaction. Roughly 80% of developers, across all experience levels, said coding felt more enjoyable when working with AI. Ultimately, while we may not have figured out yet how effective AI is for developers, they seem to want it to stay, for the time being.

The Universal Postal Union (UPU) has reported a drop of more than 80% in postal traffic to the US, following the formal end of the de minimis exemption at the end of August by the Trump administration.

The UPU has also reported that many of its members have suspended some or all postal services to the US, until a suitable solution is reached. The UPU is actively working on a solution, in the form of a landed-cost calculator available to operators.

The end of de minimis is already set to have an impact on businesses. However, the UPU is committed to encouraging operators to continue postal services in the US.

Postal Traffic Plunges Following End of De Minimis Rule

Postal traffic to the US has fallen by more than 80%, according to the Universal Postal Union (UPU). In a statement made September 6th, the UPU reported that information exchanged among postal operators through its network showed traffic had fallen 81%, compared with a week earlier.

This comes after the Trump administration ended the de minimis exemption for small packages on August 29th.

The UPU boasts a membership of 192 countries, nearly all the countries in the world, and said it had sent a letter to the US secretary of state Marco Rubio in order to express concerns about the impact of ending de minimis. The Trump administration has described the rule as a loophole for foreign businesses to evade tariffs and a way for criminals to get drugs into the US.

Postal Operators Seeking Solutions Before Continuing US Service

On top of the drop in postal traffic, the UPU also reported that 88 postal operators have suspended some or all postal services to the US, until a solution is found for the delivery of parcels valued at $800 or less. This was previously the cutoff for imported goods to escape customs charges.

“The global network saw postal traffic to the US come to a near-halt after the implementation of the new rules on August 29th 2025, which for the first time placed the burden of customs duty collection and remittance on transportation carriers or US Customs and Border Protection agency-approved qualified parties.” – Universal Postal Union, in a statement

Furthermore, the UPU said they have seen “major operational disruptions,” and stated that their members had not been given enough time or guidance to comply with the procedures, outlined in Donald Trump’s July 30th executive order.

De minimis has existed since 1938. Now, purchases that previously entered the US without needing to clear customs are subject to vetting and their origin country’s tariff rate, ranging from 10% to 50%.

Universal Postal Union Deploys Potential Solution Post-De Minimis

Despite the drop, the UPU is exploring potential solutions for postal operators to resume delivery services in the US. This consists of equipping operators with a landed-cost calculator via an API that can be plugged into retail and counter solutions, so they’ll be able to calculate and collect the required duties from customers immediately before sending packages to the US.

“The UPU has in its mission the responsibility to guarantee the free circulation of postal items over a single postal territory.” – Masahiko Metoki, UPU Director, in a statement.

As a result of the ending of de minimis, ecommerce sellers have already started to hike up the prices of their products in order to comply with new rules. Consumers could also face delays in orders further down the line. With the new regulations now fully in effect, we’ll have to watch out for the longer-term consequences.

Cyberattacks at educational facilities have spiked in recent months, according to new research. A study from Check Point Research finds that between January and July 2025, there was a 41% year-on-year (YoY) increase in illicit cyber activity.

Looking at countries across APAC, Africa, Europe, Latin and North America, the study finds that institutions in the US recorded a 67% YoY increase in cyberattacks — the steepest rise across all countries surveyed. At the same time, it consolidates the education sector’s position as the most attacked across the globe.

Cybersecurity is one of the biggest issues in the business world today. Recently, it was revealed that 78% of companies had been targeted with ransomware in the last year. Successful breaches can be catastrophic, with multiple high-profile businesses paying a high price for failing to deter attacks in recent years.

Cyberattacks Spike As Students Return to School

With school and college terms starting up again, educational institutions around the world have borne witness to a huge spike in attempted cyberattacks. According to new research from Check Point Research, between January and July 2025, the sector averaged 4,356 attacks per organization each week, comprising a 41% YoY increase.

The hardest hit region is APAC, which averaged 7,869 attacks per organization, per week. North America came in at fifth place, with a 67% increase in attacks YoY — the largest yearly increase across every region surveyed.

Check Point Research used its ThreatCloud AI platform to gather the insights, which analyzes millions of indicators of compromise (IoC) daily to build up a picture of the global cyberthreat landscape.

Education Leads Worst Hit Industries Worldwide

The data finds that education is by far the worst affected industry across the world when it comes to cyberattacks. In second place is government, which records 2,716 attacks per organization, per week on average. After that, it’s telecommunications (2,636 per organization, per week) and healthcare and medical (2,468 per organization, per week).

It is not a surprise that education has become such a target in recent years. In the post-pandemic era, it has come to be defined by a huge dependence on online platforms, with login credentials and public networks proving tantalizing attack vectors for a cybercriminals.

Not only this, but educational facilities tend to have limited cybersecurity budgets, meaning that their infrastructure is easy to bypass and students and faculty are often not well-versed in how to identify different breaches.

Cybersecurity Is A Thorn in the Business World’s Side

Eagle-eyed observers will not be surprised by the new data. Cybersecurity has emerged as one of the most urgent concerns of the business world in recent times. Seemingly every week, a major global company suffers a wide-ranging cyberattack, the consequences of which are often severe.

With the rapid development of AI, these attacks are getting harder to detect and harder to stop. Nonetheless, companies are not doing enough to insulate themselves from harm.

Businesses should be increasing their cybersecurity budgets, as well as investing in new talent and upskilling their existing employees to spot different threats. According to our own research, a shocking 98% of bosses can’t identify all the signs of a phishing attack, indicating that this is a problem endemic across the workplace.

Without serious action to avert the crisis, companies will continue to fall victim to scams, which are becoming increasingly sophisticated in nature. At this point in time, the outlook for the future is pretty bleak.

OpenAI’s ChatGPT Projects is now available at no cost to anyone with a free ChatGPT account.

This feature lets users group the various ChatGPT chats that they’ve had in the past, for faster future reference. This system of folders makes it easier to give the generative AI tool a set of custom instructions, and lets them upload project-specific files to reference.

Previously, the tool had only been available as a perk for those on paid plans.

What’s Changing for ChatGPT Users

Users can opt for three different ChatGPT pricing plans: The free plan, the ChatGPT Plus plan for $20 per month, and the ChatGPT Pro plan for $200 per month.

With the latest change, all three plans include the Projects feature, so conversations can be grouped and used together.

However, there’s still one difference between the three plans, Engadget reports: Free users will be able to upload up to five files per project for the Projects function. In contrast, Plus users can upload up to 25 and Pro users can upload up to 40.

OpenAI’s Evolving Approach to Free Features

OpenAI has been steadily increasing the premium features and model access that it offers for free users.

For example, this time two years ago, you needed to shell out for the $20 per month paid plan in order to access what was at the time OpenAI’s most advanced large language model, GPT-4. Today, you can get access to GPT-5 entirely for free (although the number of times you can use it will be limited).

Similarly, OpenAI has rolled out its Deep Research and ChatGPT Voice tools to all users, even though both of them started out as free.

Generative AI has so fully saturated culture at this point that OpenAI arguably needs to keep offering more new free abilities to keep earning headlines like the one at the top of this page.

Using ChatGPT in 2025

Granted, not all news has been good news for OpenAI.

Take one feature that they recently removed, for example: A check box that would make a users’ conversations discoverable in search engines. Many people had already accidentally enabled all their private conversations to be seen by anyone who searched for them, creating a crisis of privacy violations.

We’ve previously covered all the types of information that we strongly recommend not sharing with ChatGPT. The stuff to avoid ranges from sensitive company data to IP to financial information and passwords.

Needless to say, you should still avoid entering this data when using Projects: The newly available ability to upload and work on files might even make it easier than ever to accidentally overshare on the platform.

President Donald Trump has filed a petition asking for a review of a new federal appeals court ruling in Washington DC that found most of his trade tariffs to be illegal.

The Trump administration asked for the Supreme Court to make a new judgement by September 10th, a full month ahead of the October 14th deadline.

Without a shakeup from a new Supreme Court ruling, as much as $200 billion from Trump’s tariffs might be refunded to the importers who initially paid them, once that mid-October deadline passes.

The Current Legal Ruling

Why were the tariffs thrown out in the first place?

Because, according to both of the judges who have looked at the case so far, Trump overstepped his presidential powers when he incorrectly used a 1977 law aimed at combating national emergencies to justify issuing his tariffs. This law — the International Emergency Economic Powers Act (IEEPA) — was likely not intended to give the President unlimited authority over tariffs.

The specific tariffs that have been ruled illegal are the “liberation day” border taxes Trump introduced in early April.

Trump Says America Could Be “Unbelievably Poor” Again

During a press conference discussing this request, Trump highlighted his belief that a ruling against him would make the US “suffer,” according to a Reuters report.

“We made a deal with the European Union where they’re paying us almost a trillion dollars. And you know what? They’re happy. It’s done. These deals are all done. I guess we’d have to unwind them. […] Our country has a chance to be unbelievably rich again. It could also be unbelievably poor again. If we don’t win that case, our country is going to suffer so greatly, so greatly.” – President Trump

The accelerated schedule that Trump is asking for would allow arguments to be heard in November, with justices ruling by the end of the year.

How Is the Logistics Industry Impacted?

The beleaguered logistics industry is already dealing with additional new fees brought on by the trade war.

The de minimis exemption expired near the end of August, causing any imported packages that cost less than $800 to require duties. Major shipping companies are already passing some of these costs on to customers, with the UPS announcing recently that it would impose a new international processing fee of $2.50 per package for US imports.

Dozens of top US trade partners are already being tariffed, with countries like Brazil currently under a 50% tax. Many of these tariffs could be rolled back if the US Supreme Court agrees with the two other judges who have considered the case.

Not only are logistics professionals struggling with the ongoing driver shortage, but they are also dealing with an ageing workforce, as recent Tech.co data reveals. The industry could therefore turn to Gen Z – who are struggling to find work, in order to fill gaps behind the wheel.

The profession holds a lot of benefits for the generation, including giving them the opportunity to travel, use technology, and contribute to a purposeful and steady career path.

On top of increasing driver benefits and improving work-life balance for drivers, the logistics industry has employed a range of strategies to combat the shortage.

Why Trucking’s Ageing Workforce Needs Gen Z Drivers

According to Tech.co’s 2025 logistics report, Moving Goods With Fewer Hands, the trucking industry is currently dealing with a severe driver shortage. Not only have 63% of US freight businesses said driver recruitment and retention has stagnated or worsened in the past year, but a quarter say an ageing workforce is their biggest hiring challenge.

On the other hand, data from the US Bureau of Labor Statistics shows that 2.5 million Gen Z were unemployed in July, and 1.8 million were looking for full time work.

The driver shortage could therefore pose a great opportunity for Gen Z, as they take up roles as truck drivers that the industry is sorely lacking.

“There’s a real opportunity here to connect an industry in urgent need of new talent with a generation looking for meaningful, stable work. Trucking offers Gen Z more than just a job; it offers purpose, growth, and a real chance to shape the future of logistics.” Bianca Prieto, Editor of The Inside Lane trucking newsletter

Benefits of a Trucking Career for Gen Z

Not only would Gen Z contribute to minimizing the logistics driver shortage, but there are plenty of benefits on offer for young people joining the profession.

Travel opportunities

For Gen Z looking to travel and see the world, the trucking profession presents a great opportunity to take new journeys and see new places. In a recent webinar hosted by The Inside Lane newsletter, Lindsey Trent, President of Next Gen Trucking Association, said: “Gen Z wants to make a difference, use technology, and explore the world. Trucking checks all those boxes – we just have to tell that story better.”

Tech use

Technology is already a big part of many Gen Z lives, and being involved with the logistics industry will mean they’ll be able to use tech such as fleet management and route optimization software. Similarly, they may find working with technology more appealing than older generations who may be less tech-literate.

Trent also said: “Younger generations have tech as standard, and it has to be used to give them more of a voice – which should be encouraged by companies too.”

Purposeful and stable career

Our latest data shows that 91% of respondents agree that America’s economy depends on fast, affordable, reliable freight movement. Truck drivers are therefore critical in maintaining the US economy, and will certainly be needed in the future as freight demand increases. In fact, our data shows that 63% of freight professionals have reported an increase in demand over the past year. 

Trucking is therefore a good, stable career path for Gen Z. Likewise, Trent added that Gen Z “want to do work that matters, and they value experiences. Trucking offers that in spades. We just have to reach them early – in high schools, trade programs, and even on social media.”

How Are Trucking Companies Attracting New Talent?

According to our logistics report, increasing driver compensation and improving work-life balance (56%) are the most popular tactics being used by professionals to combat driver shortages.

Businesses are also taking on other strategies, too. This includes providing better training and development opportunities (44%), enhancing recruitment efforts (43%), and improving company culture (27%).

Ultimately, it doesn’t seem like the problem is going anywhere anytime soon. On top of tariff losses and the driver shortage, this will certainly be a defining year for the resilience of the logistics industry.

Good news! If you can’t be bothered to search the internet any more, soon you won’t have to. That’s right — there’s a new Claude Chrome extension in town.

Earlier this week, Anthropic announced that its flagship chatbot would soon be getting its own Chrome integration. This will enable users to relinquish control of the mouse and keyboard to Claude, which will be able to see everything that you’re doing in your browser and perform complex processes on your behalf.

If that’s of interest, we’ve put together a simple guide on how to get started with the new Chrome extension.

What Is the New Claude Chrome Extension?

Anthropic has revealed that a new Google Chrome extension of its Claude chatbot will soon be widely available. With the integration enabled, users can permit Claude to take over their browser and perform multistep tasks for them — with next to no human involvement required.

For example, if you’re short on time, you could ask Claude to find a restaurant in your neighborhood that serves Spanish cuisine at a reasonable price. Or alternatively, you could instruct it to find the nearest grocery store that is open. Rather than searching and cross-checking results by yourself, the extension will do it all for you.

Announcing the partnership, Anthropic issued a statement: “We’ve spent recent months connecting Claude to your calendar, documents, and many other pieces of software. The next logical step is letting Claude work directly in your browser.” In this respect, it resembles Manus AI, supposedly the world’s first fully autonomous chatbot that launched earlier this year.

How to Get the New Claude Chrome Extension

At this point, the extension is an initial test that is only available to 1,000 subscribers of Claude Max, which costs either $100 or $200 per month. Anthropic plans to gradually roll out the trial to more Max subscribers, but there’s no word on when it will be usable by customers at different tiers.

If you meet the above criteria, you can sign up for a waitlist. You’ll get taken to a signup form and asked to fill out some key details, which are as follows:

• Are you 18 years of age or older?
• What is the email address associated with your Claude account?
• Do you have a Max plan subscription?
• How do you plan to use Claude for Chrome? What workflows are you looking to automate with it?
• Can we contact you to learn more about your use cases and gather feedback on your experience?

Please note that this is the only way you can sign up for the extension pilot. Any publicly available extensions purporting to be Claude are imposters, and you should definitely avoid installing them.

Anthropic Warns of Existing Security Concerns

Part of the reason that the pilot is only available to a limited number of participants is that it has outstanding security vulnerabilities. In its press release, Anthropic noted that “prompt injection attacks” represented a genuinely possibility for people who signed up to take part in the trial.

To be specific, some of the inherent risks include malicious users accessing personal accounts and files, sharing private information, accessing financial information, and generally taking actions that the user never intended.

But if you’re prepared to take those risks, then why not sign up for the Claude extension waitlist now? Your findings will enable Anthropic to “take an important step towards a fundamentally new way to integrate AI into our lives.”