Last Week in AI

This episode covers Anthropic’s Haiku 4.

Hosts Andrey Kerenkov and John Krohn discuss OpenAI’s Sora 2 video model, Anthropic’s Claude Sonnet 4.

This episode covers OpenAI’s GPT-5 Codex upgrade, Google’s Gemini integration into Chrome, and Anthropic’s new file creation features.

This episode covers Google’s Gemini 2.

This episode covers the release of OpenAI’s GPT-5, which consolidates previous models into a single routing system, and Anthropic’s Opus 4.

This episode covers the Trump administration’s AI action plan and GitHub’s new “vibe coding” tool, Spark.

This episode covers OpenAI’s new ChatGPT Agent, Alibaba’s Kimi K2 model, and Amazon’s Kiro coding tool.

This episode analyzes xAI’s Grok 4 launch, which leads frontier benchmarks but sparked controversy over anti-Semitic outputs and alignment with Elon Musk’s views.

This episode covers Cloudflare’s new default blocking of AI scrapers and Runway’s entry into AI-generated gaming.

This episode covers Google’s Gemini CLI launch, Anthropic’s new artifact publishing features, and the ongoing trademark lawsuit between OpenAI and IO.

This episode covers Midjourney’s new video generation tool, Google’s Gemini 2.

This episode covers OpenAI’s O3 Pro release, Cursor’s 1.

This episode covers Anthropic’s voice mode, Black Forest Labs’ image editing tools, and Perplexity’s agentic features.

This episode covers major AI announcements from Google I/O 2025 and Anthropic’s Claude 4 release, highlighting advancements in coding agents, video generation, and search.

This episode covers OpenAI’s decision to retain its nonprofit structure while transitioning its for-profit arm to a public benefit corporation,…

This episode covers Anthropic’s new app integrations for Claude and OpenAI’s controversial “sycophantic” model update.

This episode covers OpenAI’s GPT-4.

This episode covers Meta’s controversial Llama 4 launch, Amazon’s Nova Act web agent, and Alibaba’s upcoming Qwen3.

This episode covers Google’s Gemini 2.

This episode covers Baidu’s cost-competitive Ernie models and OpenAI’s new audio tools and expensive O1 Pro API.

This episode covers OpenAI’s new agent-building tools and a $12B Corvi deal, signaling a shift in B2B strategy.

This episode analyzes the shift from pure pre-training to inference-time reasoning, discussing GPT-4.

This episode covers major AI releases, including OpenAI’s GPT-4.

The hosts celebrate their 200th episode and announce format changes, including retiring bios and adding listener topics.

This episode covers OpenAI’s O3 Mini and Deep Research, Google’s Gemini 2.

This episode of Last Week in AI focuses heavily on DeepSeek R1, analyzing its reinforcement learning approach to reasoning and its implications for hardware demand and US-China export controls.

This episode explores the critical intersection of AI and hardware, tracing the evolution from early custom machines to modern GPU-driven data centers.

Alessio and Swix from Latent Space guest-host this episode, covering Google and Microsoft’s new AI pricing strategies and the ongoing data licensing wars with news publishers.

This episode focuses heavily on AI hardware, detailing NVIDIA’s $3,000 personal supercomputer and its push into custom ASIC manufacturing in Taiwan to compete with Broadcom.

Host Jeremy Harris discusses OpenAI’s O3 model, highlighting its breakthrough performance on coding and reasoning benchmarks like SWE-bench and ARC-AGI.

The hosts discuss Google’s release of Gemini 2 Flash Thinking, Veo 2 video generation, and Project Mariner browser agents, alongside XAI’s Grok 2 launch.

This episode covers OpenAI’s launch of Sora and ChatGPT’s advanced voice mode, alongside Google’s Gemini 2.

This episode covers OpenAI’s new $200 ChatGPT Pro tier and Amazon’s launch of the Nova model family, alongside updates on ElevenLabs and Google’s Veo.

This episode covers the leaked OpenAI Sora video generator, discussing its quality and the controversy surrounding artist relations.

The hosts discuss reports of slowing AI progress at major labs, contrasting this with rapid advances in agentic AI and new reasoning APIs.

This episode covers new features from OpenAI and Anthropic, including speculative decoding and pricing shifts.

Host Jeremy covers recent AI news while recovering from a newborn hospital stay.

The hosts return from a break to cover Anthropic’s new agentic Claude 3.

Andrei and guest host Chris Albon discuss Adobe’s new AI video tools, YouTube’s DreamTrack expansion, and Meta AI’s global rollout.

This episode covers Meta’s MovieGen video generator, highlighting its inpainting capabilities and potential as a world simulator.

Guest co-host John Krohn joins Andrei to discuss OpenAI’s advanced voice mode, Meta’s celebrity voice integrations, and the rapid rollout of Gemini’s voice chat.

Hosts Jeremy Harris and a co-host discuss OpenAI’s release of O1, a reasoning-focused model that hides its chain-of-thought for competitive advantage.

This episode covers Amazon partnering with Anthropic to power a paid Alexa 2.

Hosts Andrei Kerenkov and Jerry Ares discuss Google’s new Gemini Gems, experimental models, and restored image generation capabilities.

This episode covers Ideogram 2’s text improvements, Google’s free Imagen 3, and faster video tools from Luma and Runway.

This episode covers the beta launch of Grok 2 with Flux image generation, OpenAI’s mysterious GPT-4.

Hosts discuss the AI startup "implosion," analyzing Google’s acqui-hire of Character.

Hosts Jeremy and Andrei discuss Apple Intelligence’s delayed iOS 18 launch, Meta’s new AI Studio for Instagram, and NVIDIA’s Blackwell chip delay.

This episode covers OpenAI’s Search GPT prototype, Google’s Gemini 1.

This episode covers OpenAI’s launch of the cheaper GPT-4o Mini and its secret "Strawberry" reasoning project.