Researchers find LLMs like ChatGPT output sensitive data even after it’s been ‘deleted’
According to the scientists, there’s no universal method by which data can be deleted from a pretrained large language model.
Archive context
Older archive item. Useful for background and entity history, but not a fresh market-moving signal.
According to the scientists, there’s no universal method by which data can be deleted from a pretrained large language model.
Why this matters
This research story adds another data point to the current market tape and is useful when read alongside nearby source coverage.
Original source
Read on CointelegraphRelated market context
Bitcoin miners escape months of distress as daily revenue surges by 78%
Bitcoin miners are emerging from months of financial pressure as rising BTC prices lift daily industry revenue by 78%. According t...
AI Could Weaken Ethereum Security as Cryptographic Risks Grow, Vitalik Buterin Warns
Buterin urged developers to prepare for potential AI vulnerabilities in both conventional and quantum-resistant cryptography, whil...
Stealing $1.5B in crypto is easy, cashing out is the trap
North Korean hackers stole around $1.5 billion from Bybit in February 2025. While the hack itself has been widely covered and anal...
Bitcoin ETFs Shed $729M in Two Days as Investors Reverse Course
Bitcoin Magazine Bitcoin ETFs Shed $729M in Two Days as Investors Reverse Course U.S. investors this week reversed course, cashing...
French authorities count 90 crypto-related crimes in seven months
The rise in crypto-related crimes in France highlights the urgent need for enhanced data security and protective measures for digi...
XRP Price Prediction: Elliott Wave Setup Points to $3 as XRP Nears a Potential $2 Breakout
XRP Price Prediction: Elliott Wave Setup Points to $3 as XRP Nears a Potential $2 Breakout XRP is approaching a critical technical...