Ohad Elhelo
@OElhelo
Followers
436
Following
397
Media
13
Statuses
124
CEO at AUI (Augmented Intelligence)
New York, USA
Joined October 2014
Bravest frenchman since Emile Zola
An AI pioneer, Yann Lecun, says his field has been led astray by Large Language Models (since they are not based on factual models of the world, but only on massive correlations and abstractions from them). As a cognitive scientist, I'm sympathetic. https://t.co/TXgWWTHxzI via
0
0
1
“We are not going to get to human-level AI just by scaling LLMs. There’s no way, absolutely no way, and whatever you can hear from some of my more adventurous colleagues, it’s not going to happen within the next two years. There’s absolutely no way in hell.” 👇🏼
88
258
2K
Augmented Intelligence, which is building neuro-symbolic AI models, raised $20M in a bridge SAFE round at a $750M valuation, bringing its total funding to ~$60M (@carlfranzen / VentureBeat) https://t.co/NnDG50Uqgh
https://t.co/ddZgfbsHBz 📫 Subscribe:
0
4
8
What language models (LM) have revealed is that the majority of the artificial intelligence community, that is supposed to be rigorous & professional, is gullible. The AI community is literally high on its own supplies. LMs have nothing to say about consciousness.
LLMs have revealed that information/computation is an essential form, with an ontological life of its own, independent of any physical substrate. Thought isn’t in matter, any more than math is in the calculator. This inverts the mind–body relationship and reframes the mind–body
44
138
851
Paging @GaryMarcus...was he right all along?!?
1
3
9
@nxthompson @OElhelo The fusion of neural networks and symbolic reasoning could be a game changer for AI development.
0
1
2
The reason it is so important for everyone to keep pretending that AGI is definitely right around the corner is that there is now over $1T of investment riding on this belief (either already expended, or committed) Current (and recent past) capex cannot be justified by current
154
377
2K
When the hype train crashes into the walls of reality, please do remember those who said LLMs can be conscious (but only during fundraising rounds), dangerous (while not able to order pizza) and cause mass unemployment ("we're writing the last piece of software") 🤡🤡🤡
0
0
7
Neuro-symbolic AI is absolutely fascinating, and will be reading this a few times to really ingest it...it's like getting a glimpse of the future. Incredible, approachable writeup by @OElhelo - I found myself wanting to take notes - this is just the beginning.
2
2
42
There’s a bit of a rift in the A.I. community over the best way to make A.I. smarter. On one side you have generative A.I., which is powered by neural networks, and on the other you have symbolic A.I., which relies on rules and reasoning. @OElhelo combined the two ideas and
3
7
21
29 months into 6 months from AI taking your job Andrej K> "I feel like the industry [...] it's trying to pretend that this [Current AI] is amazing. And it's not—it's slop"
28 months in to 6 months from AI taking your jobs * 4 months into 24 months until cursor is obsolete * 6 months into 6 months until ai writes 90% of your code (part 2, the codening)
164
272
4K
Had a great time chatting with @nxthompson on neuro-symbolic AI, Apollo-1's upcoming release and why LLM agents are ngmi
.@OElhelo’s company @Aug_Intel will soon release a model that purports to combine the fluency of generative A.I. with the accuracy of symbolic reasoning. If it works, it could be used by hundreds of millions of people. I asked him what he’s thinking about ahead of the launch.
0
2
4
I am thrilled to announce “The Most Interesting Thing in A.I.” podcast is back for a third season. Guests include: @OElhelo on the promise of neuro-symbolic A.I. @navrinasingh on building trustworthy A.I. @willahmed on how A.I. can help athletes train and recover Matt Wood on
1
2
10
Has this stealth startup finally cracked the code on enterprise AI agent reliability? Meet AUI's Apollo-1 https://t.co/vZZ5GMVGvz
0
4
6
AUI plans to launch public access to Apollo-1 later this fall, allowing companies and developers to build their own agents. The company has also announced a strategic partnership with Google Cloud.
2
2
7