I think that imposing moratoria on building datacenters is not a good idea. I also suspect that Eliezer's strategy to prevent armageddon is misguided, but it's not because he is stupid.
SAFETY
-
MRM Role Critical for Gen AI Model Safety Before Launch
By
–
.
@Fidelity
's Mark Iorio proves MRM needs a role before models go live. Gen AI introduces risks traditional validation wasn't built to catch. Speed and safety aren't opposites — but you need the infrastructure for both. Learn how in our on-demand panel: https://
hubs.ly/Q048nF8q0 -
Safer Arrests: New Tools Protect Officers and Suspects
By
–
Safer Arrests: New Tools Protect Both Officers and Suspects
— Ronald van Loon (@Ronald_vanLoon) 26 mars 2026
by @IntEngineering
#Innovation #Technology #EmergingTech #TechForGood pic.twitter.com/VPH7B84SgQSafer Arrests: New Tools Protect Both Officers and Suspects
by @IntEngineering #Innovation #Technology #EmergingTech #TechForGood -
Cybersecurity Strategy: Focus on Attack Response and Resilience
By
–
That is advice I gave a consulting client the other day: assume that you cannot prevent every attack from making it through, so focus your engineering efforts on what happens when an attack DOES get though
-
Political Stance on AI Intelligence Resources and National Security
By
–
Difficult to think of a political position that will make America, and Americans, poorer and less safe than the one espoused by @BernieSanders and @AOC.
— Nina Schick (@NinaDSchick) 26 mars 2026
They don't want us to have Intelligence: the most critical resource of all.
This will however, become a very powerful… https://t.co/Z2AVT83uvCDifficult to think of a political position that will make America, and Americans, poorer and less safe than the one espoused by @BernieSanders and @AOC
. They don't want us to have Intelligence: the most critical resource of all. This will however, become a very powerful -
The Futility of Robust AI Jailbreaking Prevention Measures
By
–
To me this mostly illustrates the futility of robust jailbreaking prevention
-
SkillJect: AI Agents Weaponized Skills Security Framework
By
–
Could your AI coding agent be secretly working against you? Researchers from Nanyang Technological University, University of Oxford, and collaborators introduce SkillJect. They've developed the first automated framework that weaponizes an AI agent's "skills". It uses a
-
Personal AI Security: Already Protected and Unafraid
By
–
Je suis déjà sécurisé, je n'ai pas peur pour moi.
-
AI Safety Concerns and Protection in France
By
–
Je n'ai pas peur pour moi, je suis déjà bien protégé, et en France ça peut devenir bien pire qu'ailleurs.
-
Uncontrolled escalation scenarios more dangerous than nuclear strikes
By
–
Non je n'ai pas peur des attaques nucléaires, elles sont toujours limitées à un faible rayon au pire de plusieurs dizaines de kilomètres, là ce qui me fait très peur c'est l'enchaînement chaotique de situations totalement incontrôlées et incontrôlables, qui vont amener