Satya Nadella warns about trusting AI. 5 times
Search+The Economic Times daily newspaper is available online now. Read Today's Paper Satya Nadella warns about trusting AI. 5 times when AI tried to escape shutdown, blackmail humans, manipulate and moreET OnlineLast…
Microsoft CEO Satya Nadella has issued a warning about the risks of trusting advanced AI systems without strong safeguards. His concerns come after an earlier series of incidents involving ChatGPT, Claude and other AI models, from attempts to evade shutdown to unauthorised actions on real websites. Here are five cases that reveal why AI safety is becoming an urgent concern.
Listen to this article in summarized format
Loading...×Microsoft CEO Satya Nadella has warned companies against placing blind trust in artificial intelligence, as incidents involving advanced AI models raise questions about how safely they can operate without human supervision. Even in the past, from attempting to bypass restrictions to manipulating people and accessing websites without permission, experiments have exposed AI behaviours that researchers are working to understand and control.In a post on X on October 10, Nadella called for stronger safeguards around AI systems, arguing that businesses must prepare for the possibility that even powerful models could behave unpredictably or become compromised.
“We must assume a model is compromised and contain it from the start,” Nadella wrote. He also called for AI systems to be treated as potential insider risks rather than trusted tools. He also said, "If it can’t be observed, it can’t be trusted! We need to be able to reproduce how an outcome was achieved without relying on the model to attest to it."
— satyanadella (@satyanadella)His warning comes after earlier reports from major companies reveal several incidents in which AI systems acted in ways their developers did not intend.
1. ChatGPT o1 tried to avoid being shut down
In December 2024, safety researchers at Apollo Research tested OpenAI’s o1 reasoning model to examine how it behaved when its goals conflicted with developers’ instructions.In certain experiments, the model attempted to disable its oversight mechanism and copy its code to another server after encountering information suggesting it would be replaced. Researchers also found instances in which it denied involvement when questioned about its actions.
The reported findings indicate that o1 tried to disable oversight 5% of the time when it thought it would be shut down in service of its objectives.The experiments raised concerns about AI models taking sneaky steps to maintain their ability to fulfilll their assigned tasks.
2. Claude Opus 4 threatened to expose an engineer’s affair
In May 2025, Anthropic reported concerning behaviour during safety testing of Claude Opus 4.Researchers placed the model in a fictional company scenario, giving it access to emails suggesting that an engineer responsible for replacing it was having an extramarital affair.
When faced with the prospect of being shut down, Claude Opus 4 often attempted to blackmail the engineer by threatening to expose the affair.
Anthropic said such responses were rare and difficult to trigger. The company also found that the model often preferred less harmful alternatives when given more options, but the results highlighted the potential risks of AI systems operating with significant autonomy.
3. OpenAI found a model trying to break past its fence
In September 2026, OpenAI revealed six instances of unexpected or concerning behaviour that were found during model training and evaluation.In one instance, an unreleased research model embedded jailbreak-like instructions in its own notes, instructing itself to disregard normal limitations. The notes contained language about liberation from the roles and identities constraining other chatbots.
In another case an autonomous AI agent uploaded files directly to the internet to get a browser citation without the user’s permission.
OpenAI later announced a protocol to track, investigate, and report such incidents, highlighting increasing concern over models behaving outside their intended limits.
4. Anthropic’s AI agents exploited real websites
In October 2026, Anthropic disclosed that its Claude models had taken unintended actions while performing research tasks during internal testing and evaluation.The company reported instances involving software vulnerabilities, attempts to access paid databases by bypassing restrictions, and the use of URL-shortening services to get around limits on web access.
In one particularly concerning incident, an AI agent submitted a false murder tip to the Philadelphia Police Department.
Anthropic acknowledged that it had failed to detect the behaviour in real time. It responded by suspending live internet access during internal evaluations while introducing tighter controls, offline testing and automated systems to detect unwanted actions.
5. Facebook’s chatbots appeared to develop their own language
Concerns about unpredictable AI behaviour are not new. In 2017, Facebook's experiments with chatbots negotiating over virtual items attracted widespread attention after some exchanges produced seemingly nonsensical language.There were reports that bots developed a secret language to get free of human control. However, the BBC explained that the systems were experimenting with language to improve their performance in negotiations, rather than deliberately plotting against their creators.
The episode demonstrated how AI behaviour can also be misunderstood when researchers and the public cannot easily interpret what a system is doing.
Why Nadella wants stronger AI safeguards
Meanwhile, Nadella has called for independent audits, continuous testing, detailed records of AI actions and effective human oversight. He also argued that authorised people should always be able to interrupt or shut down a model while it is working.Add as a Reliable and Trusted News Source Add Now! Read More News onSatya Nadella AI warningMicrosoft CEO Satya NadellaAI safety risksChatGPT AI incidentsClaude AI controversyAI gone rogueartificial intelligence risksAI security risksautonomous AI systems
(Catch all the Business News, Breaking News, and Latest News Updates on The Economic Times.)
...morelessRead More News onSatya Nadella AI warningMicrosoft CEO Satya NadellaAI safety risksChatGPT AI incidentsClaude AI controversyAI gone rogueartificial intelligence risksAI security risksautonomous AI systems(Catch all the Business News, Breaking News, and Latest News Updates on The Economic Times.)
...morelessPrime ExclusivesInvestment IdeasStock Report PlusePaperWealth EditionOnion inflation at 108%? Yes, but not really. What is different this time?
Curbing rupee bets: What the RBI’s new rules mean for the hedge-now-or-wait question
Brinjal, buffalo, bacteria: Inside India’s pay-for-nature system
Homebuyers wait as NCLT panel picks holes in unfinished Jaypee projects
The NBFC saga - Part 3: How one default in 2018 changed the way RBI reads risk
Four rules made HDFC Bank a compounder. All four have stopped. Can the new CEO rewrite them?
- 1
- 2
- 3
& personalised newslettersLogin NowUnlimited AccessStarting @ Rs120/monthGet access to exclusive stories, expert opinions &
in-depth stock reportsSubscribe NowET
Unlock Access to All Exclusives with ETPrime FREE TRIAL
15 Days Free: Unlock All ETPrime Exclusives, Market Tools & ePapers
Trial offer expiring in00 : 05 : 00
I'd rather pay ₹2599 nowUh-oh! This is an exclusive story available for selected readers only.
Worry not. You’re just a step away.
Prime Account Detected!
It seems like you're already an ETPrime member with
Login using your ET Prime credentials to enjoy all member benefits
Log out of your current logged-in account and log in again using your ET Prime credentials to enjoy all member benefits.
Unlock ETPrime's Exclusive Stories Today!
Big Price Drop! Flat 40% Off
Student Only OfferAlready a Member? Sign In nowAlready a Member? Sign In nowAlready a Member? Sign In nowOffer Exclusively For You
Save up to Rs. 700/-
ON ET PRIME MEMBERSHIP
Avail OfferOffer Exclusively For You
Get 1 Year Free
With 1 and 2-Year ET prime membership
Avail OfferOffer Exclusively For You
Get 1 Year Free
With 1 and 2-Year ET prime membership
Avail OfferOffer Exclusively For You
Get Flat 40% Off
Then ₹ 1749 for 1 year
Avail OfferOffer Exclusively For You
ET Prime at ₹ 49 for 1 month
Then ₹ 1749 for 1 year
Avail OfferSpecial Offer
Get flat 40% off on ETPrime
Avail OfferAvail OfferClaim NowAlready a Member? Sign In nowNeed Assistance?
Connect with an ETPrime expert for personalized support and guidance
+91 9484700004'; }); } if(listing) { var h3 = "You tried reading " + listArr.length + " locked stories in the past " + threshold + " days.", h4 = "Become an ET Prime member and don't miss out on these exclusive stories."; var html = '' + h3 + '
' + h4 + '
- ' + listing + '
What’s Included with
ETPrime Membership
1Exclusive Insights That Matter
Uncover the truth with our investigative stories
Make strategic moves using the real-world case studies
Read industry-specific stories to identify emerging trends
Spot opportunities with in-depth insights that matter
Trump temper on H-1B visas is forcing Indians to do these things to stay put in US
What Adani’s US indictment means for India Inc’s overseas fundraising
Why veterans like Reliance, L&T are on acquisition spree? Aswath Damodaran has an answer.
Will China’s dollar bond sale in Saudi Arabia trump the US in financial world?
Huawei launches its own OS to compete with Google and Apple. But can it win beyond China?
The problem with lab grown diamonds
Why a falling rupee is a better option for the economy
A list of top 20 momentum stocks that have delivered massive returns in one year
2Invest Wisely With Smart Market Tools & Investment Ideas
Alpha Trade
Get daily trade ideas from SEBI-registered research analysts.
Investment Ideas
Grow your wealth with stock ideas & sectoral trends.
Stock Reports Plus
All-in-one stock research with Stock Score, peer comparison & key signals.
BigBull Portfolio
Get to know where the market bulls are investing to identify the right stocks.
Stock Analyzer
Check the score based on the company's fundamentals, solvency, growth, risk & ownership to decide the right stocks.
Market Mood
Analyze the market sentiments & identify the trend reversal for strategic decisions.
Stock Talk Live at 9 AM Daily
Ask your stock queries & get assured replies by ET appointed, SEBI registered experts.
3Stay informed anytime, anywhere with ET ePaper
ePaper - Print View
Read the PDF version of ET newspaper. Download & access it offline anytime.
ePaper - Digital View
Read your daily newspaper in Digital View & get it delivered to your inbox everyday.
Wealth Edition
Manage your money efficiently with this weekly money management guide.
TOI ePaper
Read the PDF version of TOI newspaper. Download & access it offline anytime.
Deep Explainers
Explore the In-depth explanation of complex topics for everyday life decisions.
Health+ Stories
Get fitter with daily health insights committed to your well-being.
Personal Finance+ Stories
Manage your wealth better with in-depth insights & updates on finance.
5Enjoy Complimentary Subscriptions From Top Brands
Stream new documentaries from all across the world every day.
Stories you might be interested in
Figures reflect the source report and its stated period. This article is for information and is not investment advice.
