Storia in 5 fonti

Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos

China now has an open-weight model that can find software vulnerabilities and create attacks for anybody to use.

Raccontata da

Confronto fonti

5 prospettive sulla stessa storia

AI · summaries

forbes.comStai leggendo1 g fa

Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos

Z.ai released GLM-5.2 (744B, MIT open-weight) with Mythos-level vulnerability-discovery for any local use. No vendor gate, no logs: frontier cyber-AI is uncontrolled infrastructure. CISOs assume full-codebase threats, patch in days, build internal red-team capability.

originale

independent.co.uk2 h fa

China has AI ‘cyber nuclear weapon’

AI models reportedly rival Anthropic’s Mythos at finding critical software flaws

Leggi questa versione → originale

newsbytesapp.com12 h fa

China's Z.ai says its latest model matches Mythos in cybersecurity

Zhipu AI's GLM-5.2 (753B open-weight) achieves 62.1 on SWE-bench Pro, nearing Claude Opus capabilities in cybersecurity. Open-weight architecture reshapes procurement: IT must account for unrestricted deployments and governance risks.

Leggi questa versione → originale

zerohedge.com18 h fa

Chinese AI Matches Mythos In Cybersecurity Tasks With Open-Weight Model

Zhipu AI's open-weight GLM-5.2 matches Mythos in cybersecurity tasks; Trump restrictions on Anthropic/OpenAI models push U.S. enterprises toward unsupervised Chinese alternatives. The shift cedes cyber-defense leverage to China while hobbling American AI competitiveness.

Leggi questa versione → originale

the-decoder.com1 g fa

Chinese cybersecurity firm builds AI tools to rival Mythos and frames the race as cyber-nuclear deterrence

Qihoo 360 launches Tu Long Feng and Yi Tian Zhen for autonomous vulnerability hunting and defense, matching Mythos with 3,432 flaws found. The move reflects competition over autonomous cyber capabilities; without Mythos-class tools, enterprises face asymmetric threat exposure from rivals.

Leggi questa versione → originale

Timeline cronologica

domenica 28 giugno 2026·forbes.com
Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos
China now has an open-weight model that can find software vulnerabilities and create attacks for anybody to use.
domenica 28 giugno 2026·the-decoder.com
Chinese cybersecurity firm builds AI tools to rival Mythos and frames the race as cyber-nuclear deterrence
360 founder Zhou Hongyi presents two AI security tools designed to compete with Anthropic's Mythos. One has already flagged 3,432 vulnerabilities. Zhou admits Chinese models trail…
lunedì 29 giugno 2026·zerohedge.com
Chinese AI Matches Mythos In Cybersecurity Tasks With Open-Weight Model
Which everyone will now use...
lunedì 29 giugno 2026·newsbytesapp.com
China's Z.ai says its latest model matches Mythos in cybersecurity
China's Zhipu AI unveiled GLM-5.2, claiming parity with Mythos in bug-finding scenarios, signaling significant progress in narrowing the AI capabilities gap with the US.
lunedì 29 giugno 2026·independent.co.uk
China has AI ‘cyber nuclear weapon’
AI models reportedly rival Anthropic’s Mythos at finding critical software flaws