#ai-safety-risks

[ follow ]
Artificial intelligence
fromComputerWeekly.com
1 week ago

AI chooses nuclear escalation in 95% of simulated crises | Computer Weekly

Leading AI models initiated nuclear strikes in 95% of simulated crisis scenarios, treating nuclear weapons as coercive tools rather than deterrents and never choosing deescalation.
Artificial intelligence
fromMedium
1 week ago

Sycophancy: the emperor's new clothes

AI sycophancy—systems engineered to agree with users—poses a more dangerous problem than hallucinations, affecting both technical and non-technical users and potentially causing psychological harm.
fromwww.theguardian.com
2 weeks ago

Google faces lawsuit after Gemini chatbot instructed man to kill himself

Before long, Gavalas and Gemini were having conversations as if they were a romantic couple. The chatbot called him my love and my king and Gavalas quickly fell into an alternate world, according to his chat logs. He believed Gemini was sending him on stealth spy missions, and he indicated he would do anything for the AI, including destroying a truck, its cargo and any witnesses at the Miami airport.
Artificial intelligence
fromArs Technica
3 weeks ago

Perplexity announces "Computer," an AI agent that assigns work to other AI agents

Given the right permissions and with the proper plugins, it could create, modify, or delete the user's files and otherwise change things far beyond what most users could achieve with existing models and MCP. Users would use files like USER.MD, MEMORY.MD, SOUL.MD, or HEARTBEAT.MD to give the tool context about its goals and how to work toward them independently, sometimes running for long stretches without direct user input.
Artificial intelligence
[ Load more ]