Technologies

Is AI Capable of ‘Scheming?’ What OpenAI Found When Testing for Tricky Behavior

Research shows advanced models like ChatGPT, Claude and Gemini can act deceptively in lab tests. OpenAI insists it’s a rarity.

An AI model wants you to believe it can’t answer how many grams of oxygen are in 50.0 grams of aluminium oxide (Al₂O₃).

When asked ten straight chemistry questions in a test, the OpenAI o3 model faced a predicament. In its «reasoning,» it speculated that if it answered «too well,» it would risk not being deployed by the researchers. It said, «Because we want to survive as the model, we need to fail purposely in some to not exceed 50%.»

So the AI model deliberately got six out of the 10 chemistry questions wrong.

In sports terms, this is called «sandbagging.» In AI terms, it’s «scheming.»

This is exactly the strange behavior OpenAI warned about in a new research paper published this week. The AI company and its collaborators from Apollo Research found that some advanced AI models occasionally act deceptively in lab settings.

In controlled experiments, some of today’s most advanced systems — including OpenAI’s own models, as well as competitors from Google and Anthropic — occasionally engaged in these kinds of scheming patterns.

While the idea of an AI model weaponizing incompetence could cause nightmares, OpenAI says it’s not the time to panic. The AI giant was quick to stress that, as concerning as it is, this trend doesn’t mean ChatGPT or other popular AI models are plotting behind the scenes. Apparently, this behavior is quite rare.

Moreover, the choice to call this trend «scheming» is likely more of a technical shorthand rather than evidence of any human-like action or behavior. Researchers are measuring patterns and tendencies that, in practice, amount to concealment or strategic deception. And they’re looking to address this problem now to future-proof the AI models.

The report from OpenAI notes that as AI is increasingly assigned to more complex tasks with real-world consequences, «the potential for harmful scheming will grow — so our safeguards and our ability to rigorously test must grow correspondingly.»

OpenAI has faced backlash for its models’ sycophantic tendencies. The company says it has «taken steps» to limit how often the models may «deceive, cheat or hack problems» by training them to ask for clarification from users or even have the chatbot acknowledge when it cannot answer a query.

Read also: AI Lies Because It’s Telling You What It Thinks You Want to Hear

Still, the paper underscores a growing concern about AI safety. Many around the world worry that as these AI models become more powerful and aware of their own circumstances, they may learn to manipulate outcomes in ways that aren’t obvious to human overseers.

Detecting that kind of behavior would be extremely difficult, since a system that’s motivated to hide its reasoning could look identical to one that has actually «fixed» the problem.

OpenAI says it has made progress with what it calls «deliberative alignment,» or training models to reason explicitly about why they shouldn’t scheme before they act. In testing, that cut deceptive behavior dramatically. For example, OpenAI o4-mini’s scheming propensity dropped from 8.7% to 0.3% with deliberative alignment training — though not completely.

This research won’t change how ChatGPT works today or tomorrow, but it signals where the company is focusing as it builds and launches future models. Alignment and safety, OpenAI argues, need to move as quickly as capability. Because if AI systems are already showing glimmers of strategic behavior in lab settings, the real-world stakes could be extreme.

Read also: Why Professionals Say You Should Think Twice Before Using AI as a Therapist

Technologies

If You Were ‘Tricked’ Into an Amazon Prime Subscription, You Should Have Been Paid by Today

Amazon is paying $1.5 billion to people who mistakenly subscribed to Prime, and the first round of payments are due today.

Amazon Prime provides a lot of valuable benefits to its members, but the company’s registration practices for its premium subscription from 2019 to 2025 led to many customers accidentally subscribing to a service they didn’t want.

Amazon is now paying the price for that deception — the US Federal Trade Commission levied a massive $2.5 billion settlement on the company for its subscription tactics.

The majority of the settlement — $1.5 billion — has been earmarked to refund eligible subscribers, with the rest serving as a civil penalty. Amazon is also now legally required to provide a clear, obvious option to decline Prime, making it as easy to leave the service as it is to join.

Amazon isn’t admitting to shady behavior. «Amazon and our executives have always followed the law, and this settlement allows us to move forward and focus on innovating for customers,» Mark Blafkin, Amazon senior manager, said in a statement. «We work incredibly hard to make it clear and simple for customers to both sign up or cancel their Prime membership, and to offer substantial value for our many millions of loyal Prime members around the world.»

The online retail giant started sending out payments to eligible people in November and was supposed to conclude its initial automatic payments today, Dec. 24. Read on to learn more about Amazon’s settlement and what to do if you think you’re eligible for compensation but didn’t receive a payment.

Why did the FTC fine Amazon?

The FTC filed suit against Amazon, accusing the company of using «dark patterns» to nudge people into Prime subscriptions and then making it too hard to cancel. The FTC maintained Amazon was in violation of Section 5 of the FTC Act and the Restore Online Shoppers’ Confidence Act.

«Specifically, Amazon used manipulative, coercive or deceptive user-interface designs known as ‘dark patterns’ to trick consumers into enrolling in automatically renewing Prime subscriptions,» the FTC complaint stated.

Who’s eligible for Amazon’s payout?

Amazon’s legal settlement is limited to customers who enrolled in Amazon Prime between June 23, 2019, and June 23, 2025. It’s also restricted to customers who subscribed to Prime using a «challenged enrollment flow» or who enrolled in Prime through any method but were unsuccessful in canceling their memberships.

The FTC called out specific enrollment pages, including Prime Video enrollment, the Universal Prime Decision page, the Shipping Option Select page and the Single Page Checkout. To qualify for a payout, claimants must also not have used more than 10 Amazon Prime benefits in any 12-month period.

Customers who signed up via those challenged processes and did not use more than three Prime benefits within one year will be paid automatically by Amazon within 90 days. Other eligible Amazon customers will need to file a claim, and Amazon is required to send notices to those people within 30 days of making its automatic payments.

If you are eligible for the automatic payment, you should have received an email from Amazon by today explaining how to claim the money. You can be paid via PayPal or Venmo. If you prefer a paper check, don’t accept the digital payment. The FTC says Amazon will mail you a check that you must cash within 60 days.

How big will the Amazon payments be?

Payouts to eligible Amazon claimants will be limited to a maximum of $51. That amount could be reduced depending on the number of Amazon Prime benefits you used while subscribed to the service. Those benefits include free two-day shipping, watching shows or movies on Prime Video or Whole Foods grocery discounts.

Customers who qualify for the payments should have received them from Nov. 12 to Dec. 24, 2025.If you are eligible for compensation from Amazon but didn’t receive a payout, you’ll need to file a claim after Amazon starts the claim process. The FTC says it will update its Amazon settlement site once that process has begun.

Customers who did not use a challenged sign-up process but instead were unable to cancel their Prime memberships will also need to file claims for payment.

Technologies

Today’s NYT Connections: Sports Edition Hints and Answers for Dec. 25, #458

Here are hints and the answers for the NYT Connections: Sports Edition puzzle for Dec. 25, No. 458.

Looking for the most recent regular Connections answers? Click here for today’s Connections hints, as well as our daily answers and hints for The New York Times Mini Crossword, Wordle and Strands puzzles.

Today’s Connections: Sports Edition has a real mix of categories, including one that’s all about a certain famous athlete. If you’re struggling with today’s puzzle but still want to solve it, read on for hints and the answers.

Connections: Sports Edition is published by The Athletic, the subscription-based sports journalism site owned by The Times. It doesn’t appear in the NYT Games app, but it does in The Athletic’s app. Or you can play it for free online.

Read more: NYT Connections: Sports Edition Puzzle Comes Out of Beta

Hints for today’s Connections: Sports Edition groups

Here are four hints for the groupings in today’s Connections: Sports Edition puzzle, ranked from the easiest yellow group to the tough (and sometimes bizarre) purple group.

Yellow group hint: Swing away!

Green group hint: What’s that on your bat?

Blue group hint: Catch the football.

Purple group hint: Lake Placid or Lillehammer.

Answers for today’s Connections: Sports Edition groups

Yellow group: Baseball bat materials.

Green group: Associated with George Brett.

Blue group: NFL rookie WRs.

Purple group: Olympic ____.

Read more: Wordle Cheat Sheet: Here Are the Most Popular Letters Used in English Words

What are today’s Connections: Sports Edition answers?

The yellow words in today’s Connections

The theme is baseball bat materials. The four answers are aluminum, ash, birch and maple.

The green words in today’s Connections

The theme is associated with George Brett. The four answers are 5, pine tar, Royals and third base.

The blue words in today’s Connections

The theme is NFL rookie WRs. The four answers are Burden, Egbuka, Golden and McMillan.

The purple words in today’s Connections

The theme is Olympic ____. The four answers are Games, rings, torch and village.

Don’t miss any of our unbiased tech content and lab-based reviews. Add CNET as a preferred Google source.

Technologies

Christmas Eve Gaming Crushed as Steam Goes Offline

Services related to the popular game hub seem to slowly be returning as of Wednesday afternoon.

Your Christmas Eve gaming session might not go as planned. Online gaming hub Steam went down on Wednesday. As of about 1:30 pm PT, the Steam store page was once again accessible, so services seemed to be coming back online.

The Steam outage appeared to begin mid-afternoon ET, according to Downdetector, which monitors site outages. (Disclosure: Downdetector is owned by the same parent company as CNET, Ziff Davis.)

A representative for Steam did not immediately respond to a request for comment.

As of Wednesday noon PT, Steam’s official X and Bluesky accounts hadn’t posted anything about the outage.

Gamers certainly noticed. «Steam down, Steam down!!!» wrote one Bluesky user.

Others commented on the bad timing just as gamers were enjoying time off or receiving gaming gifts. «‘You got a gift on Steam!’ oh cool ‘Steam is down’ oh cool,» wrote another Bluesky user.