I think this might be a pretty severe prompt injection pattern
SAFETY
-
Commitment to Privacy Over Profit in AI Development
By
–
We are committed to thriving and surviving, whatever that takes. But we WILL NOT undermine our privacy guarantees or sell out the people who rely on us.
-
LLMs in Space: Hardening llm.c to NASA Safety Standards
By
–
Clearly LLMs must one day run in Space Step 1 we harden llm.c to pass the NASA code standards and style guides, certifying that the code is super safe, safe enough to run in Space. https://
en.wikipedia.org/wiki/The_Power
_of_10:_Rules_for_Developing_Safety-Critical_Code
… (see the linked PDF)
LLM training/inference in principle should be super -
Enterprise AI Observability and Real-Time Intervention Launch
By
–
Today we’re launching new AI observability functionality with real-time intervention for generative AI solutions, available across all AI assets and environments. This latest release delivers the tools and safeguards your organization needs to confidently build enterprise-grade
-
AI Predicting Political Orientation from Photos: Increased Surveillance Risks
By
–
[#Article] When AI Predicts Political Orientation from a Photo: Toward Increased Surveillance? https://actuia.com/actualite/quand-lia-predit-lorientation-politique-a-partir-dune-photo-vers-une-surveillance-accrue/
… #AI #artificialintelligence -
AI Safety and Cybersecurity Future Panel at RSAC
By
–
Attending #RSAC? Don't miss this keynote panel including Daniel Rohrer, VP of Product Security, NVIDIA discussing the rise of #AI safety and what it means for the future of AI and cybersecurity. https://
nvda.ws/3xYEl59 -
Open Source Library Flaws Claims Questioned on Fundraising Tour
By
–
This is fanciful cope peddled on a fundraising tour imho. OS libraries have been found to be flawed, but this story is v fishy.
-
Unverified Claims Presented as Fact: Critical Analysis
By
–
Which in this context was serious re how she was sourcing unevidenced claims she presented as fact.
-
AI Collaboration Features with Citations and Data Integration
By
–
In the coming weeks, we’ll be releasing collaboration features including citations from reliable sources to verify AI-generated claims, as well as integrations with data repositories like codebases and CRMs — all while maintaining the highest standards of security and safety. pic.twitter.com/MdpA3Uuepu
— Anthropic (@AnthropicAI) 1 mai 2024In the coming weeks, we’ll be releasing collaboration features including citations from reliable sources to verify AI-generated claims, as well as integrations with data repositories like codebases and CRMs — all while maintaining the highest standards of security and safety.
-
Face Recognition Accuracy Gaps: Vendor Hype Versus Reality
By
–
It claims to work. But the more of the face that’s covered, the less accurate. And as you must know, the hype/claims around these tech made by thirsty vendors far exceeds the tech’s real capabilities.