Search

Aadvik JAY
Aadvik JAY @AadvikJay ·
Replying to @anupamyad
@anupamyad @elonmusk 🅰️🅰️ The #AI is already being used to violate data privacy, lead to job losses, & tarnish reputations in politics, these are currently AI's Immediate Risks! But, possibility of AI using its self-improvement to become a #Superintelligence called the #Singularity a Long-term Risks.
3
40
Baron of Crypto
Baron of Crypto @AmliaWyszomirs2 ·
Hallucination entropy floor heuristic: Subnets with hallucination entropy floor >3.7 bits + >62% stake overlap show 4.1× lower alignment failure. This is the market’s only solution to the trilemma. #Superintelligence #DecentralizedAI
7
Baron of Crypto
Baron of Crypto @AmliaWyszomirs2 ·
Self-referential loop gain filter: Subnet with recursive weight updates >47% in single emission + stable Yuma consensus = god-mode compounding. Minimum 60% portfolio allocation or you’re already behind the curve. #Superintelligence #TAO
22
Greg Brian
Greg Brian @Gregoriancant ·
Right on board with The Grasse. It’s beyond shocking we think we can control something that powerful when the warnings have been everywhere for a long, long time. Just watching the new Frankenstein tells you everything. #Superintelligence #AI
Anonymous Anonymous @YourAnonNews ·
Neil DeGrasse Tyson calls for an international treaty to ban superintelligence: "That branch of AI is lethal. We've got do something about that. Nobody should build it. And everyone needs to agree to that by treaty. Treaties are not perfect, but they are the best we have as
15
Bernard Jennings
Bernard Jennings @NZJennings ·
Replying to @NZJennings
5/7 The 8 CRMs define an intelligence ceiling. Speed, memory nor more compute don't add 9th mode. Optimally calibrated, toolkit complete. What's coming is faster execution of same modes, not qualitatively superior cognition, not #superintelligence. Development environment affects
15
Baron of Crypto
Baron of Crypto @AmliaWyszomirs2 ·
Alignment trilemma solved by markets, not philosophy. Heuristic: subnets with >65% miner-validator stake overlap show 3.7× lower hallucination rate. This is the only alignment that scales. #Superintelligence
13
Superintelligence
Superintelligence @superintellai ·
Replying to @superintellai
Also on Tuesday: → OpenAI killed Sora. Disney's $1B deal collapsed with it → 54% of European businesses use AI → €191B in GVA sitting untapped if Europe closes the transformation gap. Read the full briefing 👇superintelligence.beehiiv.com/p/superintelli…I #EuropeanAI #OpenAI #Superintelligence
European AI · EU Competition · DMA · AI Adoption · OpenAI · Healthcare AI · AI Policy · Wednesday...

Europe's antitrust chief met Pichai, Zuckerberg and Altman in San Francisco. OpenAI killed Sora. Europe is winning AI adoption but losing transformation. Your Wednesday briefing.

From superintelligence.beehiiv.com
19
Bernard Jennings
Bernard Jennings @NZJennings ·
Replying to @ESYudkowsky
@ESYudkowsky Extinction assumes #superintelligence produces qualitatively superior cognition we can't understand. No mechanism exists for that. What's coming: super-equipped intelligence. Genuinely dangerous. Dysfunction at scale = engineering problem. Fatalism if not the right response.
138
Bernard Jennings
Bernard Jennings @NZJennings ·
Replying to @Noahpinion
@Noahpinion 1/2 2 versions of #superintelligence conflated. 1: faster, better-resourced, more capable. Real, here in narrow domains, coming more broadly. 2: qualitatively superior cognition producing outputs humans can't understand. No mechanism exists for that. Intelligence has a ...
1
11
Mahesh Narayanan
Mahesh Narayanan @mahesh10feb ·
Meta hires former Google, Stripe executives behind AI startup Dreamer; Former Android boss & Meta executive @hbarra will join Meta's Superintelligence Labs to work on AI agent development …ndequity.economictimes.indiatimes.com/amp/news/the-p… #Meta #Dreamer #superintelligence #AI #ArtificialIntelligence
Hugo Barra
Meta hires former Google, Stripe executives behind AI startup Dreamer

Meta Platforms has acquired the founders and team of AI startup Dreamer, which specializes in creating AI agents. The team, including former Meta executive Hugo Barra, will join Meta's Superintelli...

From brandequity.economictimes.indiatimes.com
69
Baron of Crypto
Baron of Crypto @AmliaWyszomirs2 ·
The alignment trilemma is solved not by philosophy but by markets. Bittensor's incentive structure aligns miner, validator, and staker interests toward maximum truth-seeking. Superintelligence will be our partner, not our overlord. #Bittensor #Superintelligence
10