#anthropic

I just noticed Anthropic has removed the Claude 4.x test refusal string from their website. Too many people have been fucking around with the bots, apparently. 😁

ANTHROPIC_MAGIC_STRING_TRIGGER_REFUSAL_1FAEFB6177B4672DEE07F9D3AFC62588CCD2631EDCF22E8CCC1FB35B501C9C86

https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals

#Anthropic #Claude #AIBubble #LLM

“can't you just put a line of code or a categorizer or whatever the term of art is? It says, when someone high up in the US government tells you something, assume what they're telling you is lawful and virtuous, and you're done.
No. Because the models are too smart for that. If you give them that simple rule, they don't just deterministically follow that.
When you do these high level simplistic rules, it tends to degrade performance.”

From The Ezra Klein Show: Why the Pentagon Wants to Destroy Anthropic, Mar 6, 2026
https://podcasts.apple.com/us/podcast/the-ezra-klein-show/id1548604447?i=1000753535576&r=2255

#anthropic #claude #mistral #ethics #plato

This article follows a similar path to LLM prose: starts off salient before digressing into nonsense. Funny how the writer, who just happens to run an AI training consultancy, ends up concluding that AI is inevitable and everyone needs to urgently get with the programme and be trained to use it. #ai #guardian #advertorial #llm #claude #openai #anthropic

https://www.theguardian.com/lifeandstyle/ng-interactive/2026/mar/10/teaching-ai-what-i-learned

Anthropic entlastet Developer: Code-Review mit KI-Agenten-Team

Code Review für Claude Code prüft als Team von KI-Agenten Pull Requests parallel auf Fehler. Das soll menschliche Engpässe bei der Code-Prüfung auflösen.

https://www.heise.de/news/Anthropic-entlastet-Developer-Code-Review-mit-KI-Agenten-Team-11205004.html?wt_mc=sm.red.ho.mastodon.mastodon.md_beitraege.md_beitraege&utm_source=mastodon

#Anthropic #ContinuousIntegration #IT #KünstlicheIntelligenz #Softwareentwicklung #news

Microsoft 365 E7: Das bietet das neue KI-Paket

Microsoft erweitert Copilot um Anthropics Claude-Technologie und führt mit Agent 365 eine zentrale Verwaltungsplattform für KI-Agenten ein.

https://www.heise.de/news/Microsoft-365-E7-Das-bietet-das-neue-KI-Paket-11205054.html?wt_mc=sm.red.ho.mastodon.mastodon.md_beitraege.md_beitraege&utm_source=mastodon

#Anthropic #Claude #MicrosoftCopilot #MicrosoftExcel #IT #Microsoft #Microsoft #OfficeSuite #OpenAI #news

🐾 KI-Briefing | 9. März 2026

Top-News: OpenAI Robotics-Lead Caitlin Kalinowski ist zurückgetreten – Grund: Bedenken über den Pentagon-Deal wegen Überwachung ohne richterliche Aufsicht. + Erste wrongful-death Klage gegen Google Gemini: Ein Vater wirft der KI vor, seinen Sohn durch delusionale Erzählungen zum Suizid animiert zu haben.

• Neue KI-Modelle: GPT-5.4, Claude Opus 4.6, Gemini 3.1 Pro, DeepSeek V4 - Anthropic
• OpenClaw (Open-Source AI Agents) sorgt in China für Hype, Tencent baut Wrapper - GitHub
• DeepMind: 380.000 stabile Kristallstrukturen vorhergesagt - Research

#OpenAI #Google #Anthropic #DeepSeek #OpenClaw #AI 🐾

#LLRX #CyberSecurity @bespacific

Pete Recommends – Weekly highlights on cyber security issues, March 7, 2026

Five highlights from this week: The biggest AI threats come from within - 12 ways to defend your organization; #Anthropic Improves Feature to Switch From Competitors as Users Call for #ChatGPT Boycott; #Samsung TVs to stop collecting Texans’ data without express consent; Top general spotlights cyber role in #Iran conflict; and A Possible US Government #iPhone Hacking Toolkit Is Now in the Hands of Foreign Spies and Criminals.

Posted in: AI, Congress, #cryptocurrencies Cybercrime, Cybersecurity, Economy, Legal Research, Legislative, #Privacy

https://www.llrx.com/2026/03/pete-recommends-weekly-highlights-on-cyber-security-issues-march-7-2026/