Cybersecurity

concepts · 16 notes linked

Related: Anthropic · Zvi Mowshowitz · AI Policy · Openai · AI Safety · Claude · Us Government · Amazon

Notes

• A Complete Breakdown of the Claude Mythos 1 Leak and Features — Leaked Claude Mythos 1 capabilities including 69% Exploit Bench score
• AI #171: False Flag — Weekly roundup; OpenAI-linked PAC false-flag scandal
• Black Hat 2023 Review: LLMs Everywhere | DNSFilter — Black Hat 2023 LLM applications in security; best use leverages embeddings not chat interfaces
• Chinese hackers steal chip designs from major Dutch semiconductor company — perps lurked for over two years to steal NXP's chipmaking IP — Chimera group stole NXP semiconductor IP undetected for 2.5 years
• Crooks' Mistaken Bet on Encrypted Phones — How law enforcement cracked EncroChat and Sky ECC encrypted criminal phone networks
• Cyber Certifications Have Failed. How to Build and Prove Cyber Skills — Traditional cyber certifications fail; continuous exercising builds real resilience
• Fable #6: The Return of the King — Anthropic Fable 5 restored after government export-control blip
• GPT-5.6: The System Card — Review of OpenAI GPT-5.6 Sol/Terra/Luna system card
• New XLoader botnet uses probability theory to hide its servers — XLoader malware uses law of large numbers to hide C2 servers
• News | SpiderOak — SpiderOak cybersecurity news covering zero-trust and UAS security
• The Once And Future Fable #2 — Reconstructing the 24 hours of the Fable takedown
• The Once And Future Fable #3: Fix This Code — Debunking the Fable "jailbreak," export-control fallout
• The Once And Future Fable #4 — Fable takedown aftermath, cyber-defense, restoration odds
• The Once And Future Fable #5 — AI-policy roundup on ad hoc licensing regime and open weights
• WSJ Article Claiming China Has Matched Anthropic Is Obvious Nonsense — Debunking WSJ headline that China matched Mythos on cyber
• White House Will Ad Hoc Decide Who Can Individually Access GPT-5.6 — Critique of ad hoc White House frontier-model access regime