PulseAugur
LIVE 12:27:16
commentary · [1 source] ·
0
commentary

Why Google failed to make GPT-3 + why Multimodal Agents are the path to AGI — with David Luan of Adept

David Luan, a prominent figure in AI who previously worked at OpenAI and led LLM efforts at Google, discussed why Google missed the opportunity to develop GPT-3. He explained that Google's internal resource allocation system, the Brain Credit Marketplace, hindered large-scale projects by requiring consensus from numerous colleagues. Luan contrasted this with OpenAI's focused approach, which allowed them to take bigger risks and achieve breakthroughs. He also touched upon his current work at Adept, focusing on AI agents capable of performing human-like computer tasks as a path to AGI. AI

Summary written by None from 1 source. How we write summaries →

RANK_REASON The item is an interview discussing past AI development decisions and future approaches, rather than a new release or significant event.

Read on Latent Space Podcast →

Why Google failed to make GPT-3 + why Multimodal Agents are the path to AGI — with David Luan of Adept

COVERAGE [1]

  1. Latent Space Podcast TIER_1 · Latent.Space ·

    Why Google failed to make GPT-3 + why Multimodal Agents are the path to AGI — with David Luan of Adept

    <p><em>Our next SF event is </em><a href="https://lu.ma/aiux" target="_blank"><em>AI UX 2024</em></a><em> - let’s see the new frontier for UX since </em><a href="https://www.latent.space/p/build-ai-ux" target="_blank"><em>last year</em></a><em>! </em></p><p><em>Last call: we are …