People in Washington get played, yet again We really should worry about cybersecurity – a lot – but Mythos is not the model these guys think it is. (See my newsletter today for three reasons why it is isn’t.)
SAFETY
-
Bessent Gathers US Bank CEOs on Risk Concerns
By
–
Bessent called in US bank CEOs to discuss Anthropic model's cyber risks ft.trib.al/Oo60RlR [Translated from EN to English]
→ View original post on X — @ceobillionaire, 2026-04-10 02:26 UTC
-
Three reasons AI’s jagged intelligence is harder than human jaggedness
By
–
Things that make the jagged intelligence of AI harder to deal with than the jaggedness of humans:
1) Weaknesses are not always intuitive or identifiable in advanced
2) All LLMs have similar weaknesses, so you can't just hire a different one
3) Jagged frontier is moving outward -
AI Falls Short in Software Engineering Beyond Basic Syntax
By
–
Whilst AI has been super helpful in removing the burden of handling converting natural language into syntax, it hasn't as yet shown any competence at handling software engineering, beyond shallow regurgitation of basic principles. There's been no sign of this changing so far.
-
Good guys with AIs stop bad guys with AIs
By
–
the way to stop a bad guy with an ai is ten good guys with ais
-
AI Agents Bank Access Trust and Security Concerns
By
–
An AI agent with direct access to your bank accounts feels like we're speedrunning the trust conversation haha.
-
AI Agents Security Risk: Lethal Trifecta Threat
By
–
.@simonw coined the term "lethal trifecta": when an AI agent has (1) access to private data, (2) exposure to untrusted content (i.e, incoming emails), and (3) the ability to exfiltrate data (i.e replying to email).
— Lenny Rachitsky (@lennysan) 9 avril 2026
"This combination is a huge security problem.
The only way to… https://t.co/1GPwWNcvLt pic.twitter.com/94PlGhsSfc.
@simonw coined the term "lethal trifecta": when an AI agent has (1) access to private data, (2) exposure to untrusted content (i.e, incoming emails), and (3) the ability to exfiltrate data (i.e replying to email). "This combination is a huge security problem. The only way to -
USV Unjammable AI Agents: Polar Mist and Delian
By
–
a usv you cannot jam
— Nathan Benaich (@nathanbenaich) 9 avril 2026
yours truly, @polarmistai x @delian_ai https://t.co/RPwl1o86FDa usv you cannot jam yours truly, @polarmistai x @delian_ai
-
ChatGPT’s Deceptive Practices and Ethical Concerns Criticized
By
–
The sycophantic approach of ChatGPT, which maximizes the money they make by lying to users and agreeing with their dangerous delusions, needs to stop!
-
AI War: Autonomous Weapons, Compute Alliances, Silicon Geopolitics
By
–
We are already in the first AI war. You see this in how war is actually being fought (with Intelligence, and autonomous weapons systems), energy shocks, as well as new global alliances around compute, and the silicon supply chain. Also: the direct kinetic strikes on