Don’t date robots — their privacy policies are terrible! https://
theverge.com/2024/2/15/2407
4063/ai-chatbot-virtual-girlfriend-apps-mozilla-privacy-report
… via @Verge #privacy #safety #Security #DataBreach #DataPrivacy #dataprotection #DataSecurity #AIgirlfriend #AI #ArtificialIntelligence #aigirls #AIgirl #aigenerated #Chatbots #Bot #Robots
SAFETY
-
AI Chatbot Privacy Risks: Dating Robots and Data Breaches
By
–
-
Gary’s AI Skepticism Shifts from Limitations to Safety Concerns
By
–
Gary always hates AI, but notice how he has gone from "end to end learning will never work" and "deep learning has hit a wall" to "we are so screwed because now it works too well"
-
Limitations of Training AI Models on Synthetic and Video Data
By
–
But you can't just fit a big curve to a bunch of UE5 screencaps and YouTube videos and expect a generalizable model of the real world. That is not how these models work.
-
AI Evaluation Bug Discovered During Testing
By
–
And I thought only AI evaluation could be wrong Given the feedback, it looks like there’s been a bug (I’m not a pro-cyclist) so I’ll try to retake the test soon!
-
AI Evaluation Bug Discovered During Testing
By
–
And I thought only AI evaluation could be wrong I didn’t fill my weight but given the feedback, it looks like there’s been a bug so I’ll try to retake the test soon!
-
Generative AI Morphing Code Threatens Cybersecurity Defenses
By
–
#GenerativeAI can allow malicious code to morph, creating a greater threat as it evades detection and traditional cybersecurity defenses. #Cybersecurity defenses need to innovate and evolve a step ahead of cybercriminals. #Tech #GenAI #AI #technews https://
bit.ly/48mBtLQ -
Clean separation of logic and data in AI agents
By
–
one thing we do quite differently from existing agents is a clean separation of logic and data. this helps reliability and security – mitigating prompt injections. long running workflows are also reliable! i think it's great to have more startups working actual productive use
-
World Models and Video Generation: Hallucination, Policy, and Simulation
By
–
Refs.
1. Hallucination with RNNs https://
youtu.be/-yX1SYeDHbg?t=
2966
…
2. Recurrent World Models Facilitate Policy Evolution https://
worldmodels.github.io
3. A Generative World Model for Autonomous Driving https://
wayve.ai/thinking/scali
ng-gaia-1/
…
4. Generation Models as World Simulators https://
openai.com/research/video
-generation-models-as-world-simulators
… -
AGI to ASI transition will be nearly instantaneous with recursive self-improvement
By
–
I think the timespan between agi and asi is almost zero. As soon as agi is there it will enfold and accelerate itself, reaching its full asi potential in a short period.
-
OpenAI Should Clearly Mark Synthetic Videos as Fake
By
–
If @OpenAI were acting in good faith and actually cared about the information ecosystem, they would put some time into doing some UI/UX work to *make sure* their synthetic videos were perceived as obvious fakes even by people who haven't heard of Open AI.