Čínská AI se v kybernetické bezpečnosti vyrovná modelu Mythos od Anthropic

35 sources
  • Podle společnosti Graphistry se model s otevřenými vahami GLM-5.2 od Zhipu AI vyrovnal proprietárním modelům od společností Anthropic a OpenAI v benchmarku kybernetické bezpečnosti, a to při zhruba polovičních nákladech.
  • Společnost 360 Security Technology představila nástroje, které její zakladatel označil za „čínskou verzi Mythos“, a tvrdí, že identifikovaly přes 3 400 softwarových chyb, což částečně potvrdily i čínské úřady.
  • Společnost Anthropic samostatně obvinila Alibabu z provozování 25 000 falešných účtů za účelem získávání dat z modelu Claude v rámci operace, kterou označila za významný „destilační útok“.
Sources (35)
  1. 1 GLM 5.2 Open Model: Beats Sonnet, Matches Opus in Cyber Evals www.graphistry.com
  2. 2 Experts say it rivals Claude Mythos from Anthropic - YouTube www.youtube.com
  3. 3 China's 360 Security Technology unveils AI models for vulnerability ... www.scworld.com
  4. 4 360 Unveils AI Tools To Match Anthropic's Mythos | Let's Data Science letsdatascience.com
  5. 5 Anthropic accuses Alibaba of obtaining illicit access to Claude www.ft.com
  6. 6 Anthropic Claims Alibaba Ran 'Brazen' Campaign to Access Its ... www.wsj.com
  7. 7 Источник: https://www.wsj.com/tech/ai/chinese-ai-anthropic-mythos ... www.threads.com
  8. 8 China Has Matched Anthropic in Cybersecurity, Resetting AI Race x.com
  9. 9 Zhipu's GLM 5.2 Surges Past Google Models as US Export Ban Hobbles Anthropic mlq.ai
  10. 10 Radoslav Krehlik's Post - LinkedIn www.linkedin.com
  11. 11 What Is GLM-5.2? Inside Z.ai's 744B-Parameter Agentic AI Model www.reddit.com
  12. 12 GLM-5.2: The ChatGPT Moment for Local AI - Linas's Newsletter linas.substack.com
  13. 13 Anthropic Tells White House Alibaba Used 25,000 Fake Accounts to ... www.citybiz.co
  14. 14 Anthropic Says Alibaba 'Illicitly' Accessing AI Model - Bloomberg.com www.bloomberg.com
  15. 15 Anthropic says Alibaba-linked operators used fake accounts to flood ... www.instagram.com
  16. 16 U.S. Anthropic ban opens door for open-source AI, particularly from ... fortune.com
  17. 17 China's new open-source model accelerates AI hacking threat - Axios www.axios.com
  18. 18 GLM-5.2 Strengthens China's Open-stack Hedge Against U.S. hackernoon.com
  19. 19 360 Unveils AI Cybersecurity Tools to Compete with Anthropic's ... www.kucoin.com
  20. 20 What is this slop? Do they actually mean GLM 5.2-Cyber (non ... x.com
  21. 21 Chinese cybersecurity company 360 unveils new AI tools ... www.facebook.com
  22. 22 Anthropic Said Alibaba Illegally Accessed Claude incrypted.com
  23. 23 The Hacker Sent by Anthropic to Calm the Government's Nerves ... www.wsj.com
  24. 24 Anthropic Set to Preview Powerful 'Mythos' Model to Ward Off AI ... www.wsj.com
  25. 25 What You Can Do About Anthropic's Mythos Superhacker AI - WSJ www.wsj.com
  26. 26 Anthropic claims that China's Alibaba used ... - Tom's Hardware www.tomshardware.com
  27. 27 Mythos Rewires the Bug-Bounty Industry - WSJ www.wsj.com
  28. 28 25000 fake accounts hit Claude. Anthropic blames Alibaba. - Threads www.threads.com
  29. 29 Cybersecurity Braces for AI 'Bugmaggedon' - The Journal. - WSJ www.wsj.com
  30. 30 Anthropic's Bombshell Letter To U.S. Officials Claims Alibaba 'Illicitly ... www.youtube.com
  31. 31 AI Shifts Cyber's Hardest Problem From Finding Flaws to Fixing Them www.wsj.com
  32. 32 DeepSeek Scrambles U.S.-China Tech War - WSJ www.wsj.com
  33. 33 The AI Price War Is Here, Piling Pressure on OpenAI and Anthropic www.wsj.com
  34. 34 China Has Matched Anthropic in Cybersecurity, Resetting AI Race www.reddit.com
  35. 35 “China Has Matched Anthropic in Cybersecurity, Resetting AI Race ... x.com

Leave a Reply

Your email address will not be published. Required fields are marked *

Stay informed and not overwhelmed, subscribe now!