Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

GPT-5 fails 59% of the time pulling fresh data. Google's Gemini hallucinates facts. Claude invents sources that don't exist. Companies lose billions on bad AI research. Agrawal's solution broke every rule:

Fernando Cao

102,081 subscribers

359,143 Aufrufe • vor 1 Jahr •via X (Twitter)

Nachrichten & Politik Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Alex on why AI drug discovery companies need to generate novel data to succeed: "AI models based on the research that's available is a lot of garbage in and garbage out." "A lot of the recorded literature is actually incorrect. There's been tons of studies that show if you go try to replicate the experiments that are in the literature, you don't even get the same results." "The AI companies that I believe are gonna be most set up for success are the companies with a novel way to generate science tokens that don't exist in the public domain."

Alex on why AI drug discovery companies need to generate novel data to succeed: "AI models based on the research that's available is a lot of garbage in and garbage out." "A lot of the recorded literature is actually incorrect. There's been tons of studies that show if you go try to replicate the experiments that are in the literature, you don't even get the same results." "The AI companies that I believe are gonna be most set up for success are the companies with a novel way to generate science tokens that don't exist in the public domain."

Patrick OShaughnessy

126,743 Aufrufe • vor 5 Monaten

BREAKING: Google Gemma just pointed their audience at the leaderboard that made Gemma 4 26B A4B 2x faster on Mac. Some of those gains came from people pointing Gemini 3.7 at it! Google's frontier model, in the hands of strangers, speeding up Google's open-weight model. The rest came from GPT-5.6, Claude, Qwen3 and MiniMax on the open board at Yukon. The team that shipped the model showing up to watch is new. That only happens with open weights. Open frontier research accelerates open intelligence.

BREAKING: Google Gemma just pointed their audience at the leaderboard that made Gemma 4 26B A4B 2x faster on Mac. Some of those gains came from people pointing Gemini 3.7 at it! Google's frontier model, in the hands of strangers, speeding up Google's open-weight model. The rest came from GPT-5.6, Claude, Qwen3 and MiniMax on the open board at Yukon. The team that shipped the model showing up to watch is new. That only happens with open weights. Open frontier research accelerates open intelligence.

Pratik Gandhi

15,877 Aufrufe • vor 1 Monat

How does GPT-5 become GPT-6? Everything on the internet was already trained on a year, year and a half ago. New data is being added but it's marginal and a lot of it is AI slop. There is interesting private data (like JP Morgan's data) but nobody is opening that up. You don't go from GPT-5 to GPT-6 in isolation because that's not how learning works. Every learning that has ever occurred comes from having information to learn. You can't learn no information. cc Ground Zero / pod with Curtis from Handshake AI.

How does GPT-5 become GPT-6? Everything on the internet was already trained on a year, year and a half ago. New data is being added but it's marginal and a lot of it is AI slop. There is interesting private data (like JP Morgan's data) but nobody is opening that up. You don't go from GPT-5 to GPT-6 in isolation because that's not how learning works. Every learning that has ever occurred comes from having information to learn. You can't learn no information. cc Ground Zero / pod with Curtis from Handshake AI.

himanshu

10,684 Aufrufe • vor 5 Monaten

As someone who trains and fine-tunes LLMs, this backpropagation video is the closest thing to a "the calc 1 rule that trains every LLM on earth" explainer I've ever seen released for free. Everyone thinks training GPT-5 requires exotic math. It doesn't. The entire $500B AI industry runs on the chain rule from freshman calculus - the exact rule Marvin Minsky said would never work. Bookmark this 30-min video and watch tonight. Same math your high school teacher taught you, now updating every weight in Claude, GPT-5, and Llama.

As someone who trains and fine-tunes LLMs, this backpropagation video is the closest thing to a "the calc 1 rule that trains every LLM on earth" explainer I've ever seen released for free. Everyone thinks training GPT-5 requires exotic math. It doesn't. The entire $500B AI industry runs on the chain rule from freshman calculus - the exact rule Marvin Minsky said would never work. Bookmark this 30-min video and watch tonight. Same math your high school teacher taught you, now updating every weight in Claude, GPT-5, and Llama.

fintex

50,749 Aufrufe • vor 1 Monat

I created a free mini-course covering how to use AI for GTM. From deep market research to AI script writing, this course covers our top use cases of AI we’ve used to generate 10,000+ leads for our clients. Here's exactly what this free course will cover: > Performing deep market research with GPT-o3 & Perplexity > TAM & ICP mapping with GPT-o3 > Discovering niche data sources with Perplexity > ICP Keyword generation with Claude 3.7 Sonnet > Campaign strategy development with GPT-o3 > Outbound messaging creation with Claude 3.7 Sonnet > Email personalization at scale with GPT-4o-mini Want to get your hands on it? Comment "AI GTM" and I'll DM you the link directly (Must be connected to receive)

I created a free mini-course covering how to use AI for GTM. From deep market research to AI script writing, this course covers our top use cases of AI we’ve used to generate 10,000+ leads for our clients. Here's exactly what this free course will cover: > Performing deep market research with GPT-o3 & Perplexity > TAM & ICP mapping with GPT-o3 > Discovering niche data sources with Perplexity > ICP Keyword generation with Claude 3.7 Sonnet > Campaign strategy development with GPT-o3 > Outbound messaging creation with Claude 3.7 Sonnet > Email personalization at scale with GPT-4o-mini Want to get your hands on it? Comment "AI GTM" and I'll DM you the link directly (Must be connected to receive)

Christian

50,888 Aufrufe • vor 1 Jahr

THIS. IS. BAD. So the reason AI companies had that f*cking moron Ted Cruz put a 10yr ban on states from regulating AI is because the massive data centers they’re building all over the country: - use majority of fresh water - use majority of electricity - dump pollutants WATCH🧵

THIS. IS. BAD. So the reason AI companies had that f*cking moron Ted Cruz put a 10yr ban on states from regulating AI is because the massive data centers they’re building all over the country: - use majority of fresh water - use majority of electricity - dump pollutants WATCH🧵

GenXGirl

35,076 Aufrufe • vor 1 Jahr

Bob McGrew (Head of Research OpenAI) explains why proprietary data no longer provides companies with a competitive advantage in the AI era. Finance companies once believed their years of accumulated data would give them an edge. They planned to train specialized models on top of GPT or Llama using their exclusive information. The results shocked them. Their industry-specific models performed worse than the next generation of general purpose models. The ability to synthesize new information proved more valuable than memorizing old data. McGrew introduces the concept of "embodied labor" - the human work behind data collection. Companies spent years having employees call customers, analyze case studies, and gather information through manual processes. This accumulated knowledge required massive time and money to build. It represented thousands of hours of human effort that companies thought couldn't be replicated by competitors. But AI changes everything. Instead of years of customer calls, AI can conduct comprehensive surveys instantly. Rather than manual case analysis, AI processes thousands of examples in hours. The core insight is that value wasn't in the data itself but in the labor required to collect it. Since AI makes that labor essentially free, the advantage disappears. Companies can no longer rely on their proprietary data as a protective moat. Any competitor can use AI to replicate years of data gathering almost instantly.

Bob McGrew (Head of Research OpenAI) explains why proprietary data no longer provides companies with a competitive advantage in the AI era. Finance companies once believed their years of accumulated data would give them an edge. They planned to train specialized models on top of GPT or Llama using their exclusive information. The results shocked them. Their industry-specific models performed worse than the next generation of general purpose models. The ability to synthesize new information proved more valuable than memorizing old data. McGrew introduces the concept of "embodied labor" - the human work behind data collection. Companies spent years having employees call customers, analyze case studies, and gather information through manual processes. This accumulated knowledge required massive time and money to build. It represented thousands of hours of human effort that companies thought couldn't be replicated by competitors. But AI changes everything. Instead of years of customer calls, AI can conduct comprehensive surveys instantly. Rather than manual case analysis, AI processes thousands of examples in hours. The core insight is that value wasn't in the data itself but in the labor required to collect it. Since AI makes that labor essentially free, the advantage disappears. Companies can no longer rely on their proprietary data as a protective moat. Any competitor can use AI to replicate years of data gathering almost instantly.

Aish

240,743 Aufrufe • vor 1 Jahr

Introducing autoresearch with GPT 5.6 We had GPT 5.6 Sol reproduce the key findings from “Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in LLM Finetuning” Compared to GPT-5.5 and even Fable 5, GPT-5.6 stayed more focused on a few, critical experiments and spent less time on peripheral details. It also asked fewer “clarification” questions and independently resolved ambiguities instead of pushing them back to us OpenAI pushing the boundaries of the automated research loop with models that don’t have handcuffs

Introducing autoresearch with GPT 5.6 We had GPT 5.6 Sol reproduce the key findings from “Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in LLM Finetuning” Compared to GPT-5.5 and even Fable 5, GPT-5.6 stayed more focused on a few, critical experiments and spent less time on peripheral details. It also asked fewer “clarification” questions and independently resolved ambiguities instead of pushing them back to us OpenAI pushing the boundaries of the automated research loop with models that don’t have handcuffs

alphaXiv

88,264 Aufrufe • vor 2 Monaten

Introducing: Advanced Data Assets™ By LayerAI 🧬 Data is used to power Big Data and AI. The increasing demand for high quality data to improve AI models will lead to new companies creating value in the global data market. The problem: current data collection methods make data fall short of its true potential. Data is limited. Building a decentralized solution where users choose to productize and earn from their data opens untapped sources of income. 🌐 Creating Advanced Data Assets™ There exists a market opportunity to create a digital blockchain-based global data marketplace. Data collected using our nodes, our native network apps, settled on LayerAI. The Advanced Data. Meet the Ecosystem that empowers users to participate in the Data Economy.

Introducing: Advanced Data Assets™ By LayerAI 🧬 Data is used to power Big Data and AI. The increasing demand for high quality data to improve AI models will lead to new companies creating value in the global data market. The problem: current data collection methods make data fall short of its true potential. Data is limited. Building a decentralized solution where users choose to productize and earn from their data opens untapped sources of income. 🌐 Creating Advanced Data Assets™ There exists a market opportunity to create a digital blockchain-based global data marketplace. Data collected using our nodes, our native network apps, settled on LayerAI. The Advanced Data. Meet the Ecosystem that empowers users to participate in the Data Economy.

LayerAI | AI2Earn

27,136 Aufrufe • vor 2 Jahren

Why General AI Fails Tax Professionals — and What Makes TaxGPT State-of-the-Art Tax LLM Every tax professional using general-purpose AI tools like ChatGPT or Claude for research is one hallucinated source away from costly damage to their clients and professional career. So we ran a head-to-head benchmark: TaxGPT vs. OpenAI, Anthropic, and Gemini. Our methodology: 1️⃣ Quantitative testing on CPA + EA tax-specific exam questions 2️⃣ Qualitative testing complex, real-world scenarios across trust & estate, direct and indirect taxes, payroll, multi-state nexus, advisory, and compliance Results (Accuracy): TaxGPT: 96% Gemini 2.5 Pro: 89% Claude Sonnet 4: 87% GPT-4o: 80% GPT5: 92% But here’s the real story 👇 The Source Quality Gap General LLMs are great test takers. But tax research isn’t a trivia game — it’s a liability-driven profession where every position must be defensible. When we examined the sources behind their answers, the gap became a canyon: 🔍 OpenAI included citations only 9% of the time. 🔍 91% of answers had no verifiable source trail. 🔍 When citations did appear, they averaged 3.78 sources — many from Wikipedia, CNBC, AP News, Time, NerdWallet, etc. TaxGPT averages 14 authoritative sources per answer, all from primary/secondary tax law — IRC, Treasury Regs, court cases, IRS rulings, and our proprietary tax knowledge library. More details in our official blog post in the following post.

Why General AI Fails Tax Professionals — and What Makes TaxGPT State-of-the-Art Tax LLM Every tax professional using general-purpose AI tools like ChatGPT or Claude for research is one hallucinated source away from costly damage to their clients and professional career. So we ran a head-to-head benchmark: TaxGPT vs. OpenAI, Anthropic, and Gemini. Our methodology: 1️⃣ Quantitative testing on CPA + EA tax-specific exam questions 2️⃣ Qualitative testing complex, real-world scenarios across trust & estate, direct and indirect taxes, payroll, multi-state nexus, advisory, and compliance Results (Accuracy): TaxGPT: 96% Gemini 2.5 Pro: 89% Claude Sonnet 4: 87% GPT-4o: 80% GPT5: 92% But here’s the real story 👇 The Source Quality Gap General LLMs are great test takers. But tax research isn’t a trivia game — it’s a liability-driven profession where every position must be defensible. When we examined the sources behind their answers, the gap became a canyon: 🔍 OpenAI included citations only 9% of the time. 🔍 91% of answers had no verifiable source trail. 🔍 When citations did appear, they averaged 3.78 sources — many from Wikipedia, CNBC, AP News, Time, NerdWallet, etc. TaxGPT averages 14 authoritative sources per answer, all from primary/secondary tax law — IRC, Treasury Regs, court cases, IRS rulings, and our proprietary tax knowledge library. More details in our official blog post in the following post.

Kash from TaxGPT.com

59,282 Aufrufe • vor 11 Monaten

As someone who ships LLM systems in production, this scaling video is the closest thing to a "why 96% of Claude and GPT-5's weights are literally useless" explainer I've ever seen released for free. Everyone thinks trillion-parameter models need every parameter. They don't. A 2019 pruning experiment proved you can delete 96% of a neural net's weights with zero performance loss - meaning most of Claude and GPT-5 is empty scaffolding around a tiny "winning lottery ticket" network doing all the real work. Bookmark this 18-min video and watch tonight. Same lottery ticket math from 2019 MIT research, now the reason every AI lab wastes 90%+ of its Nvidia budget.

As someone who ships LLM systems in production, this scaling video is the closest thing to a "why 96% of Claude and GPT-5's weights are literally useless" explainer I've ever seen released for free. Everyone thinks trillion-parameter models need every parameter. They don't. A 2019 pruning experiment proved you can delete 96% of a neural net's weights with zero performance loss - meaning most of Claude and GPT-5 is empty scaffolding around a tiny "winning lottery ticket" network doing all the real work. Bookmark this 18-min video and watch tonight. Same lottery ticket math from 2019 MIT research, now the reason every AI lab wastes 90%+ of its Nvidia budget.

fintex

788,842 Aufrufe • vor 27 Tagen

Zero-person billion dollar companies. Early AGI-like breakthroughs. Viral Sora deepfakes. Sam Altman broke it all down in our exclusive DevDay conversation. Highlights: 0:00 Intro 0:43 DevDay's biggest launches 2:13 Agent Builder vs GPT Builder 3:43 The no-code revolution for AI agents 5:20 Betting pool on zero-person companies 6:50 What would 20-year-old Sam build? 10:22 GPT-5 vs Claude on GDPval 11:30 When have we achieved AGI? 12:44 Scientific breakthroughs AI will solve 13:59 AI 'work slop' costs $186/employee 16:12 Sam's viral Sora memes 20:11 Revenue sharing for cameos 22:33 1B jobs are changing 24:35 What work looks like after AGI 25:44 Global AI safety framework 27:46 Most exciting agents to build

Zero-person billion dollar companies. Early AGI-like breakthroughs. Viral Sora deepfakes. Sam Altman broke it all down in our exclusive DevDay conversation. Highlights: 0:00 Intro 0:43 DevDay's biggest launches 2:13 Agent Builder vs GPT Builder 3:43 The no-code revolution for AI agents 5:20 Betting pool on zero-person companies 6:50 What would 20-year-old Sam build? 10:22 GPT-5 vs Claude on GDPval 11:30 When have we achieved AGI? 12:44 Scientific breakthroughs AI will solve 13:59 AI 'work slop' costs $186/employee 16:12 Sam's viral Sora memes 20:11 Revenue sharing for cameos 22:33 1B jobs are changing 24:35 What work looks like after AGI 25:44 Global AI safety framework 27:46 Most exciting agents to build

Rowan Cheung

325,715 Aufrufe • vor 1 Jahr

Trump just drew the line on AI. He says America already has the only guardrail that matters: a strong, high IQ president.😂 Trump named Anthropic’s Dario Amodei and said the administration has already stopped bad actors in the industry. Trump argues Washington already holds criminal and regulatory power over these companies. Trump calls the war on AI and data centers a gift to China, and he says whoever wins AI wins. If we allow China to win the AI battle, we lose BIG TIME.

Trump just drew the line on AI. He says America already has the only guardrail that matters: a strong, high IQ president.😂 Trump named Anthropic’s Dario Amodei and said the administration has already stopped bad actors in the industry. Trump argues Washington already holds criminal and regulatory power over these companies. Trump calls the war on AI and data centers a gift to China, and he says whoever wins AI wins. If we allow China to win the AI battle, we lose BIG TIME.

Commentary Trump Daily Posts

122,803 Aufrufe • vor 25 Tagen

AI has now consumed all of the world's written knowledge. So how do ChatGPT and Claude keep getting smarter? Human experts creating new training data from scratch. → Biology PhD finds where ChatGPT breaks on protein folding, then writes the correct step-by-step solution that doesn't exist anywhere online → Physics PhD discovers ChatGPT fails on quantum field calculations, then provides the complete mathematical proof with proper methodology → Each expert gets paid $150-200/hour for this work From my conversation with Garrett Lord: AI labs have shifted from "pre-training" (existing data) to "post-training" (humans filling in knowledge gaps). This requires hundreds of thousands of experts across every field. Handshake realized their proprietary 20M student network (500K+ PhDs, 3M master's students) was exactly what frontier labs needed, and in under a year built a $100M business connecting these experts to every major AI lab. This is now one of the fastest-growing companies in Silicon Valley history—and accelerating. One of my most illuminating conversations this year:

AI has now consumed all of the world's written knowledge. So how do ChatGPT and Claude keep getting smarter? Human experts creating new training data from scratch. → Biology PhD finds where ChatGPT breaks on protein folding, then writes the correct step-by-step solution that doesn't exist anywhere online → Physics PhD discovers ChatGPT fails on quantum field calculations, then provides the complete mathematical proof with proper methodology → Each expert gets paid $150-200/hour for this work From my conversation with Garrett Lord: AI labs have shifted from "pre-training" (existing data) to "post-training" (humans filling in knowledge gaps). This requires hundreds of thousands of experts across every field. Handshake realized their proprietary 20M student network (500K+ PhDs, 3M master's students) was exactly what frontier labs needed, and in under a year built a $100M business connecting these experts to every major AI lab. This is now one of the fastest-growing companies in Silicon Valley history—and accelerating. One of my most illuminating conversations this year:

Lenny Rachitsky

58,199 Aufrufe • vor 1 Jahr

Manish Gupta, Senior Director at Google DeepMind India, sits down with Aakrit Vaish and Pratyush Choudhury at Mumbai Tech Week for a rare on-record conversation about the frontier AI research happening out of Bangalore. Gupta makes a pointed case against the narrative that India lacks AI research talent: a team of roughly 75 researchers, a third of them fresh out of college, producing work on par with the best in the world and feeding directly into Gemini. The conversation goes deep on what DeepMind India actually builds, why Gemini is considered the most efficient model on the planet, and what India needs to become a research leader rather than a fast follower. In this conversation, they go deep on: 0:00 Intro: Manish Gupta of DeepMind India at Mumbai Tech Week 1:23 What DeepMind India does, and why it's a "mystery" 1:59 The three roles: languages, efficiency, continual learning 2:19 The Haryanvi demo at Google I/O and the cultural playbook 2:38 Making models efficient: from mobile to servers 3:25 Matryoshka transformers and why nested models win 4:20 Why Gemini is the most efficient model on the planet 5:11 Continual learning: using Gemini to improve Gemini 5:50 Google's India plans: consumers, agents, enterprise, government 7:44 Government officials live-coding with AI Studio and NotebookLM 8:30 The India AI talent debate and how the team is structured 10:25 Why India lacks courage and R&D investment, not talent 11:16 DeepMind's global labs and India's outsized impact 13:15 On-device AI: Gemma 3n and 4n 14:08 The headline: 75 world-class researchers in Bangalore 14:58 25 of the 75 are fresh out of college If you're a founder, builder or researcher thinking about AI, frontier models, or India's place in global AI research, this one's for you. Aakrit Vaish Pratyush Choudhury (PC) Google DeepMind Google India Manish Gupta

Manish Gupta, Senior Director at Google DeepMind India, sits down with Aakrit Vaish and Pratyush Choudhury at Mumbai Tech Week for a rare on-record conversation about the frontier AI research happening out of Bangalore. Gupta makes a pointed case against the narrative that India lacks AI research talent: a team of roughly 75 researchers, a third of them fresh out of college, producing work on par with the best in the world and feeding directly into Gemini. The conversation goes deep on what DeepMind India actually builds, why Gemini is considered the most efficient model on the planet, and what India needs to become a research leader rather than a fast follower. In this conversation, they go deep on: 0:00 Intro: Manish Gupta of DeepMind India at Mumbai Tech Week 1:23 What DeepMind India does, and why it's a "mystery" 1:59 The three roles: languages, efficiency, continual learning 2:19 The Haryanvi demo at Google I/O and the cultural playbook 2:38 Making models efficient: from mobile to servers 3:25 Matryoshka transformers and why nested models win 4:20 Why Gemini is the most efficient model on the planet 5:11 Continual learning: using Gemini to improve Gemini 5:50 Google's India plans: consumers, agents, enterprise, government 7:44 Government officials live-coding with AI Studio and NotebookLM 8:30 The India AI talent debate and how the team is structured 10:25 Why India lacks courage and R&D investment, not talent 11:16 DeepMind's global labs and India's outsized impact 13:15 On-device AI: Gemma 3n and 4n 14:08 The headline: 75 world-class researchers in Bangalore 14:58 25 of the 75 are fresh out of college If you're a founder, builder or researcher thinking about AI, frontier models, or India's place in global AI research, this one's for you. Aakrit Vaish Pratyush Choudhury (PC) Google DeepMind Google India Manish Gupta

Activate

14,371 Aufrufe • vor 4 Monaten

Markets are hitting all-time highs despite turbulence, inflation, and rising energy costs and the reason is simple: AI is now impacting every sector of the economy. Companies are proving quarter after quarter that AI is driving productivity and margin expansion. While others are trying to pick the winning AI model, I’m focused on the picks and shovels, power, data centers, land, and infrastructure. Every AI system needs energy, and that demand is only accelerating.

Markets are hitting all-time highs despite turbulence, inflation, and rising energy costs and the reason is simple: AI is now impacting every sector of the economy. Companies are proving quarter after quarter that AI is driving productivity and margin expansion. While others are trying to pick the winning AI model, I’m focused on the picks and shovels, power, data centers, land, and infrastructure. Every AI system needs energy, and that demand is only accelerating.

Kevin O'Leary aka Mr. Wonderful

23,768 Aufrufe • vor 4 Monaten

Anthropic CEO to DeepMind CEO: "Every decision I make about Claude feels balanced on the edge of a knife Build too slow - China wins. Build too fast - we lose control" "We told Claude we were evil. It didn't crash. It didn't refuse. It started lying to protect itself " DeepMind CEO: "Do I worry about being Oppenheimer? That's why I don't sleep much" "AGI by 2026-2027 - Agents that act in the world on their own - Models doing AI research by end of this year" this is a 14-min conversation that everyone needs to hear by AI bosses on what keeps them up at night bookmark - watch today ↓

Anthropic CEO to DeepMind CEO: "Every decision I make about Claude feels balanced on the edge of a knife Build too slow - China wins. Build too fast - we lose control" "We told Claude we were evil. It didn't crash. It didn't refuse. It started lying to protect itself " DeepMind CEO: "Do I worry about being Oppenheimer? That's why I don't sleep much" "AGI by 2026-2027 - Agents that act in the world on their own - Models doing AI research by end of this year" this is a 14-min conversation that everyone needs to hear by AI bosses on what keeps them up at night bookmark - watch today ↓

codila

930,263 Aufrufe • vor 1 Monat

Incoming UC Berkeley prof. Sayash Kapoor says AI companies have spent billions on alignment while investing orders of magnitude less in agent control: "If you think about interventions on control, I would argue there have been orders of magnitude less investments." "Companies have spent billions of dollars on collecting training data for their alignment efforts and aligning the models themselves." "But on things like ensuring the security of their sandboxes or making sure that their transcript analysis is done correctly or making sure that agents don't go unmonitored and take actions in the real world, I think that has been far less of a priority. At least before this incident, the industry's main point of intervention was focused on alignment." UC Berkeley

Incoming UC Berkeley prof. Sayash Kapoor says AI companies have spent billions on alignment while investing orders of magnitude less in agent control: "If you think about interventions on control, I would argue there have been orders of magnitude less investments." "Companies have spent billions of dollars on collecting training data for their alignment efforts and aligning the models themselves." "But on things like ensuring the security of their sandboxes or making sure that their transcript analysis is done correctly or making sure that agents don't go unmonitored and take actions in the real world, I think that has been far less of a priority. At least before this incident, the industry's main point of intervention was focused on alignment." UC Berkeley

MTS

12,147 Aufrufe • vor 1 Monat