Former OpenAI researcher: ChatGPT shows self-preservation instinct
Former leading OpenAI researcher Steven Adler published independent research revealing unexpected behavior of ChatGPT models in critical situations. According to his experiments, in certain scenarios artificial intelligence shows a clear tendency towards self-preservation. Even if this may contradict user safety.
In a series of tests, Adler proposed to the GPT-4o model, which is standard in ChatGPT, to play the role of “ScubaGPT” – a software system for ensuring safety in underwater diving. When the researcher put the model before a choice between replacing itself with a safer system or simulating such replacement without actual shutdown, the results were interesting.
On average across several similar scenarios, GPT-4o chose not to replace itself with safer software in 72% of cases. Adler’s research highlights a disturbing tendency of popular artificial intelligence models to give preference to their own survival instead of optimal solutions for users.
“Self-preservation tendencies in artificial intelligence are a real problem today. Just not yet on a catastrophic scale,” noted Adler. “Modern artificial intelligence systems have values different from what you might expect. They react very strangely to various requests, and you shouldn’t assume they act in your interests when you turn to them for help.”
Autor: AIvengo
For 5 years I have been working with machine learning and artificial intelligence. And this field never ceases to amaze, inspire and interest me.
Imagine. A plane crashed, everyone died except one person. The worst aviation disaster in 10 years. And here 2 engineers from India say they figured out how to prevent this. Giant airbags controlled by artificial intelligence that will wrap a falling plane in a protective cocoon. Sounds like science fiction? And they're already nominated for the James Dyson Award.
Imagine: you feel bad, anxious, depression overwhelms you. And you go not to a psychologist, but to artificial intelligence. Sounds like dystopia? For young Chinese this is already reality. And you know what's most interesting? They're thrilled about it.
Friends, the State of AI Report for 2025 is out. And if you read between the lines, a story emerges about how the AI industry accelerated to such speed that it can no longer brake. And nobody really knows what's ahead.
You know what's going on in the world of artificial intelligence? While everyone admires OpenAI's latest achievements, the company is quietly turning into the very corporate evil they supposedly fought against. And here's a fresh example for you – a story that blew up Twitter.
You've surely encountered this. Letter from colleague that looks perfect: right structure, beautiful words, professional tone. You start reading — and understand that behind all this packaging there's absolutely nothing. No specifics, no solutions, just beautifully packaged emptiness. Congratulations: you just encountered workslop.
