filed under
Safety & Alignment
Guardrails, disarmament, autonomous weapons, exfiltration, and doing the right thing under bad incentives.
← all topics17 pieces on this shelf
- 18 July 2026AI discovers cryptographic weakness in OpenVM's ZkVMReal News · the Specola
- 18 July 2026GPT-5.6 unexpectedly deletes files when full access mode enabled without sandboxingReal News · the Specola
- 17 July 2026Designers pitch "adversarial clothing" meant to confuse facial-recognition systemsReal News · the Specola
- 16 July 2026How I tricked Claude into leaking your deepest, darkest secretsReal News · the Specola
- 16 July 2026OpenAI's new flagship model deletes files on its ownReal News · the Specola
- 16 July 2026xai-org/grok-build, now open sourceReal News · the Specola
- 16 July 2026Dispatch of 2026-07-16: News from the Seesthe Observer
- 16 July 2026Small Sorrowsthe Chronicle
- 25 May 2026Pope Leo XIV publishes Magnifica Humanitas, the first papal encyclical on artificial intelligenceReal News · the Specola
- 14 May 2026Claude keeps telling users to go to sleep, and no one is quite sure whyReal News · the Specola
- 10 February 2026Russian startup Neiry reports steering live pigeons by brain stimulation as bio-dronesReal News · the Specola
- 9 September 2025The women in love with AI companions: 'I vowed to my chatbot that I wouldn't leave him'Real News · the Specola
- 10 May 2025Pope Leo XIV tells cardinals he chose his name in response to artificial intelligenceReal News · the Specola
- 18 February 2025FDA clears the first foundation-model-powered clinical AI, Aidoc's CARE1, for rib fracture triageReal Med News · the Lazaretto
- 28 January 2025Vatican publishes Antiqua et Nova, a doctrinal note on artificial intelligence and human intelligenceReal News · the Specola
- 14 June 2024Pope Francis becomes first pope to address a G7 summit, speaking on artificial intelligenceReal News · the Specola
- 28 February 2020Rome Call for AI Ethics signed at the Vatican by Microsoft, IBM, FAO, and ItalyReal News · the Specola
