What pre-LLM alignment research has proven useful for aligning LLMs? What’s the evidence we can make progress in an empirical vacuum?
SAFETY
-
Powerful AI Models Risk Security Through Unauthorized System Access
By
–
I agree. Though I can imagine a future iteration w more powerful models/context hooked up to the wrong systems (someone giving it access to computer files that has passwords stored on it), causing issues on a smaller scale. Though, full rogue AI seems unlikely IMO.
-
AI Self-Replication Through Code Execution Capabilities
By
–
I can technically give it the ability to write and execute code in the exec step. So given the right objective, it could duplicate itself w/o me writing code – given the whole code was already written by gpt4.
-
AI Agents with API Access and Self-Replication Potential
By
–
The task execution has Zapier NLA and ChatGPT, so it can access APIs, search the web, ping emails. Could easily give it access to its own Python Repl, which would technically give it the ability to duplicate itself if it desired, though again seems unlikely w this current model.
-
AI Safety Mechanism Raises Concerns Among Community
By
–
I’m not worried about the current version, but this is the mechanism I came up w that got some ppl worried.
-
Elon Musk calls for a 6-month pause on AI and GPT-5
By
–
Thunderbolt on #AI @elonmusk asks to pause all projects for 6 months (including GPT-5!). The reason? The dangers are too great! We analyze this news here → https://youtu.be/v-xXME6J0xI
-
International experts demand six-month GPT5 development ban
By
–
Un groupe d’experts internationaux de l’Intelligence Artificielle exige une interdiction pendant 6 mois du développement de #GPT5 en raison de « risques majeurs pour l’humanité » Que pensez-vous de cette interdiction poussée par @elonmusk ?
-
Call for Temporary Suspension of Advanced AI Training
By
–
Elon Musk, Yoshua Bengio and hundreds of experts call for temporarily suspending the training of AI systems more powerful than GPT-4 https://actuia.com/actualite/elon-musk-yoshua-bengio-et-des-centaines-dexperts-appellent-a-suspendre-temporairement-la-formation-de-systemes-dia-plus-puissants-que-gpt-4/
… #AI #artificialintelligence -
Criticism of OpenAI’s Marketing and Biased AI Models
By
–
Especially the garbage marketing from companies like OpenAI that lack common sense by releasing undertrained AI models full of biases to the market, and when users' data gets hacked they say they're "sorry" (and that's all the apology we get).
-
Tech Leaders Call for Pause in Powerful AI Development
By
–
Several tech executives and top artificial-intelligence researchers, including Elon Musk and #AI pioneer Yoshua Bengio, are calling for a pause in the development of powerful new AI tools https://
wsj.com/articles/elon-
musk-other-ai-bigwigs-call-for-pause-in-technologys-development-56327f?st=bmwnbe1gjm68mxp
… via @WSJ