in case you think adding more rules in the system prompt will somehow fix this problem
SAFETY
-
Any Rule Can Be Bypassed With New Methods
By
–
sure you could make the prompt even more complex to try to combat this but the point here is that practically any rule you give it will always be able to be bypassed by some new method
-
Securing Billions of AI Agents: Proactive Defense Strategies
By
–
Any ideas on how to increase security (proactive defense) in a world where there are billions of AI agents (AutoGPT, BabyAGI) running in apps and servers, and we don’t know what they are talking about? Some kind of antivirus against AI agents?
-
AI Researcher Marek Rosa Discusses Autonomous Agent Safety
By
–
Moj novy rozhovor vo Forbes SK: "Sledujeme, či sa naša umelá inteligencia nesnaží ovládnuť svet, hovorí výskumník Marek Rosa" https://
forbes.sk/marek-rosa-tre
nuje-svoje-ai-v-hrach-jeho-autonomni-agenti-maju-svetove-prvenstvo/
… 1/2 -
Generative AI Universes: The End of Physical Reality
By
–
Yes, we will get generative universes, sci-fi worlds, and all kinds of wonders. We won't even need reality anymore
-
AI Innovation Halts: Ethics vs Unforeseen Consequences Debate
By
–
Experts debate whether halting AI innovation is ethical or are we risking unforeseen consequences? To read more about a thought-provoking discussion on the future of AI, visit: https://
bit.ly/3UsFSHy @ylecun @random_walker @timnitGebru @elonmusk @stevewoz @harari_yuval -
RLHF LLMs Challenge: Prompt Engineering Evil Characters
By
–
Yes, it’s challenging to make RLHF trained LLMs to act evil, e.g. if you want a psychopathic character to act and talk like one. What usually happen is that they talk like nice people, compliment, have empathy. But you can prompt engineer them to act closer to their intended
-
Driverless Cruise vehicles spotted in SF neighborhood during daylight hours
By
–
3 driverless Cruises in my SF neighborhood in the last 60 secs. Couldn’t tell if they had passengers. My app won’t let me call one till later tonight. Saw them driverless in daylight hours last week. Need to go on ride with them to see how they’ve improved in last 12 months.
-
AI’s Inhuman Advantage in Modern Warfare
By
–
#AI’s Inhuman Advantage in War
#RuleoftheRobots
@WarOnTheRocks https://
warontherocks.com/2023/04/ais-in
human-advantage/
… -
OpenAI ChatML Claims to Mitigate Injection Attacks
By
–
correct, although OpenAI does claim ChatML should help mitigate injections halfway down this page https://
github.com/openai/openai-
python/blob/main/chatml.md
…