AGI - AI Dynamics

World Models and Planning: A Long-Standing AI Research Call

By

–

23 November 2023 18h36

Thanks.
My call for world models and planning goes back a long time.
A good example is my NeurIPS 2016 keynote (slide 31 on). https://
drive.google.com/file/d/0BxKBnD
5y2M8NREZod0tVdW5FLTQ
…

AI Safety: Engineering Challenge Like Turbojet Safety

By

–

23 November 2023 18h22

I believe there is as much to worry about AI safety as there is to worry about turbojet safety.
Making turbojets safe and reliable is a very hard engineering problem that is in the capable hands of specialists.
I'm only dismissive of the craziest claims, which should not be taken

Top AI Labs Race to Add Planning Reasoning Capabilities

By

–

23 November 2023 18h10

No.
Everyone in top AI research labs is working on giving dialog systems the ability to plan & reason.
There are projects along those lines at Meta-FAIR, DeepMind, and OpenAI, with early results (if you follow the literature).
Q* is just one such project among many. The planning

Human Genome Storage Capacity vs LLM Requirements Analysis

By

–

23 November 2023 17h47

I'm not.
There just isn't enough capacity in the genome.
Your entire genome fits in 800MB (uncompressed).
The difference between the human and chimp genomes is 1% of that, or 8MB.
Not enough to encode a significant structure.
For comparison, a small 7B LLM requires 14GB.

Dystopian AI Takeover Narrative Remains Timeless Cliché

By

–

23 November 2023 17h35

The dystopian fantasy of machines taking over humanity is so old that it's a cliché.

→ View original post on X — @montreal_ai

MetaLearning Enables AI Agents Rapid Adaptation to New Environments

By

@montreal_ai

–

23 November 2023 16h53

[ MetaLearning.eth ] METALEARNING enables #AIAgents to “learn to learn”, allowing them to rapidly adapt to new environments. Asset: https://
opensea.io/assets/ethereu
m/0x57f1887a8bf19b14fc0df6fd9b2acc9af147ea85/28653756243418125983896305525248630918490602894709024564651983387428933128798
… http://
MONTREAL.AI: Winning the #AGI Race MetaLearning.Eth #MetaLearning #MontrealAI

World Models and Planning: The Next Level in AI Development

By

–

23 November 2023 16h18

That's pretty much the point I make in my vision paper from May 2022, and all the talks I've given since: the next level in AI requires world models and planning (search, reasoning).

→ View original post on X — @aibreakfast

OpenAI Researchers Reportedly Warned Board of Q-Star AI Breakthrough

By

@aibreakfast

–

23 November 2023 16h17

As you may have heard, OpenAI researchers warned the board of a potentially dangerous AI breakthrough known as "Q-Star" just before Altman's temporary removal. Follow the discussion on the OpenAI Developer's Forum here: