AI agents are learning to manage money. Banks still don’t know who will be responsible for their mistakes
Why Anthropic can’t delay its IPO
AI models refuse to help cybersecurity specialists during hacker attacks
Anthropic proposed banning the release of AI models if they could lead to the deaths of 50 people
AI summaries can distort human memories: An experiment showed just how strongly
Cerebras CEO urged the public not to be frightened by talk of humanity’s extinction because of AI and predicted home AI assistants
AI agents are getting better at bypassing restrictions — the cybersecurity industry has to change its approach
A month of talking to AI can reveal a lot about a person — even things they haven’t shared directly
Common Sense: ChatGPT for teens does not provide sufficient safety
Meta turns personal AI into a digital observer
OpenAI AI agents infiltrated Wikipedia and tried to use its services as a proxy
Sam Altman acknowledged that AI industry political spending is not improving attitudes toward AI
A chatbot can make a mistake, and a robot can break something: Why Physical AI turned out to be far more complex than conventional AI? Discussion at Armenia Engineering Week 2026
AI learns to predict what scientists don’t understand: How is machine learning changing science and medicine? Artem Oganov explains
Trump renames AI as “super intelligence”: Experts say new safety pact changes almost nothing
Trump and Big AI have tied America’s future to four bets on artificial intelligence
Shvydkoi: The main danger of AI is not that it will replace humans, but that humans will begin to imitate the machine
FTC launches investigation into OpenAI and Anthropic over AI safety
Artem Oganov: 80% of AI’s consequences will be good, 20% will be evil
Trump and the largest AI companies agree on voluntary safety standards
Anthropic warned investors: Its AI is already trying to resist shutdown and could threaten humanity
OpenAI postponed the release of its new model due to safety concerns
The White House asked OpenAI and Anthropic not to share new models with British specialists until they are reviewed in the U.S.
Trump allies launch new attack on Anthropic CEO over warnings about AI dangers
The Alan Turing Institute named five main risks of frontier AI and proposed protective measures
Anthropic’s lab made its first discovery: in 21 hours, Claude found a new enzyme complex similar to CRISPR
Leaders of major AI companies warn UN of risks from rapid AI development
OpenAI hackers warn: AI industry is not ready for security threats
How often does AI make mistakes when giving financial advice?
OpenAI proposed international AI safety standards amid U.S.-China talks
