DESK TECHNOLOGY & RESEARCH · 5 MIN
Rebuilt: how AI agents should remember – and what held up in our test
How does an AI agent remember what a conversation was about without believing everything it is told? The automation vendor n8n has published five recommendations on this.
CREDIBILITY 3/5 – Partly robust
-
Microsoft Research intern reports ML forecast of space-weather risk for US substations
-
Heise: Calif.io shows AI-assisted, reboot-proof iOS 27 jailbreak at Hexacon
-
DeepMind releases AlphaGenome Atlas, a precomputed map of DNA variant predictions
-
DeepMind says it watermarked AI-designed proteins in lab tests
-
Microsoft Research says its Magnetar zkVM proves programs 10 to 39 times faster
-
Opinion authors say youth social media bans are easily circumvented
-
Google DeepMind and GovTech Singapore describe on-device scam detection model
-
Google DeepMind details Gemini 4 Argon safeguards, citing mostly internal tests