Last Week in AI podcast | Listen online for free

Available Episodes

5 of 242

#203 - Gemini Image Gen, Ascend 910C, Gemma 3, Gemini Robotics
Our 203rd episode with a summary and discussion of last week's big AI news! Recorded on 03/14/2025 Hosted by Andrey Kurenkov and Jeremie Harris. Feel free to email us your questions and feedback at [email protected] and/or [email protected] Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. Join our Discord here! https://discord.gg/nTyezGSKwP In this episode: OpenAI's new 'deep research' feature has raised concerns about cybersecurity and the potential misuse of AI models for bio-weapons and autonomous capabilities, prompting new safety and governance measures. Google's extensive $3 billion investment in Anthropic is revealed, aligning with their AI strategy and reinforcing the importance of multiple technology partnerships. Huawei's advancements in the AI chip industry are highlighted, with significant progress in producing chips comparable to Nvidia's H100, despite export control challenges. China's recent directive discourages AI executives from traveling to the US, reflecting heightened security concerns and potentially signaling a more adversarial stance in the AI race. Timestamps + Links: (00:00:00) Intro / Banter (00:01:30) News Preview Tools & Apps (00:02:30) OpenAI launches new tools to help businesses build AI agents (00:08:50) You can now test Gemini 2.0 Flash’s native image output (00:13:32) Waymo is now offering 24/7 robotaxi rides in Silicon Valley (00:17:19) Moonvalley releases a video generator it claims was trained on licensed content (00:21:11) Snap introduces AI Video Lenses powered by its in-house generative model (00:23:37) Sudowrite Launches Muse AI Model That Can Generate Narrative-Driven Fiction Applications & Business (00:27:48) In another chess move with Microsoft, OpenAI is pouring $12B into CoreWeave (00:30:54) Huawei’s Ascend 910C Takes on NVIDIA as China’s AI Race Heats Up: More Alleged Details (00:36:26) Huawei reportedly acquired two million Ascend 910 AI chips from TSMC last year through shell companies (00:40:27) Inside Google’s Investment in the A.I. Start-Up Anthropic (00:43:26) Meta is reportedly testing in-house chips for AI training (00:46:48) Elon Musk's xAI buys 1 million sq ft site for second Memphis data center (00:50:02) Superintelligence startup Reflection AI launches with $130M in funding Projects & Open Source (00:53:11) Google calls Gemma 3 the most powerful AI model you can run on one GPU (00:58:18) Sesame, the startup behind the viral virtual assistant Maya, releases its base AI model (01:01:13) Reka AI Open Sourced Reka Flash 3: A 21B General-Purpose Reasoning Model that was Trained from Scratch (01:04:19) Open-Sora 2.0: Training a Commercial-Level Video Generation Model in $200k Research & Advancements (01:06:25) Google’s Gemini Robotics AI Model Reaches Into the Physical World (01:14:33) Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning (01:23:29) Deep Research System Card (01:29:50) Claude 3.7 Sonnet System Card Policy & Safety (01:33:24) Detecting misbehavior in frontier reasoning models (01:39:30) China tells its AI leaders to avoid US travel over security concerns, WSJ reports (01:43:48) Outro
--------
1:46:23
#202 - Qwen-32B, Anthropic's $3.5 billion, LLM Cognitive Behaviors
Our 202nd episode with a summary and discussion of last week's big AI news! Recorded on 03/07/2025 Hosted by Andrey Kurenkov and Jeremie Harris. Feel free to email us your questions and feedback at [email protected] and/or [email protected] Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. Join our Discord here! https://discord.gg/nTyezGSKwP In this episode: Alibaba released Qwen-32B, their latest reasoning model, on par with leading models like DeepMind’s R1. Anthropic raised $3.5 billion in a funding round, valuing the company at $61.5 billion, solidifying its position as a key competitor to OpenAI. DeepMind introduced BigBench Extra Hard, a more challenging benchmark to evaluate the reasoning capabilities of large language models. Reinforcement Learning pioneers Andrew Bartow and Rich Sutton were awarded the prestigious Turing Award for their contributions to the field. Timestamps + Links: cle picks: (00:00:00) Intro / Banter (00:01:41) Episode Preview (00:02:50) GPT-4.5 Discussion (00:14:13) Alibaba’s New QwQ 32B Model is as Good as DeepSeek-R1 ; Outperforms OpenAI’s o1-mini (00:21:29) With Alexa Plus, Amazon finally reinvents its best product (00:26:08) Another DeepSeek moment? General AI agent Manus shows ability to handle complex tasks (00:29:14) Microsoft’s new Dragon Copilot is an AI assistant for healthcare (00:32:24) Mistral’s new OCR API turns any PDF document into an AI-ready Markdown file (00:33:19) A.I. Start-Up Anthropic Closes Deal That Values It at $61.5 Billion (00:35:49) Nvidia-Backed CoreWeave Files for IPO, Shows Growing Revenue (00:38:05) Waymo and Uber's Austin robotaxi expansion begins today (00:38:54) UK competition watchdog drops Microsoft-OpenAI probe (00:41:17) Scale AI announces multimillion-dollar defense deal, a major step in U.S. military automation (00:44:43) DeepSeek Open Source Week: A Complete Summary (00:45:25) DeepSeek AI Releases DualPipe: A Bidirectional Pipeline Parallelism Algorithm for Computation-Communication Overlap in V3/R1 Training (00:53:00) Physical Intelligence open-sources Pi0 robotics foundation model (00:54:23) BIG-Bench Extra Hard (00:56:10) Cognitive Behaviors that Enable Self-Improving Reasoners (01:01:49) The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems (01:05:32) Pioneers of Reinforcement Learning Win the Turing Award (01:06:56) OpenAI launches $50M grant program to help fund academic research (01:07:25) The Nuclear-Level Risk of Superintelligent AI (01:13:34) METR’s GPT-4.5 pre-deployment evaluations (01:17:16) Chinese buyers are getting Nvidia Blackwell chips despite US export controls
--------
1:19:52
#201 - GPT 4.5, Sonnet 3.7, Grok 3, Phi 4
Our 201st episode with a summary and discussion of last week's big AI news! Recorded on 03/02/2025 Join our brand new Discord here! https://discord.gg/nTyezGSKwP Hosted by Andrey Kurenkov and guest host Sharon Zhou Feel free to email us your questions and feedback at [email protected] and/or [email protected] Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. In this episode: - The release of GPT-4.5 from OpenAI, Anthropic's Claude 3.7, and Grok 3 from XAI, comparing their features, costs, and capabilities. - Discussion on new tools and applications including Sesame's new voice assistant and Google's AI coding assistant, Gemini Code Assist, highlighting their unique benefits. - OpenAI's continued user growth despite competition, pricing models for Google's text-to-video platform, and HP acquiring and shutting down Humane's AI pin. - Insights into new research on alignment and specification gaming in LLMs, including papers on fine-tuning causing broad misalignment and Google's multi-agent system for scientific collaboration. Timestamps + Links: (00:00:00) Intro / Banter (00:01:36) News Preview Tools & Apps (00:02:33) OpenAI announces GPT-4.5, warns it’s not a frontier AI model (00:07:22) Anthropic launches a new AI model that ‘thinks’ as long as you want (00:11:14) New Grok 3 release tops LLM leaderboards (00:16:43) Sesame is the first voice assistant I’ve ever wanted to talk to more than once (00:18:30) Google launches a free AI coding assistant with very high usage caps (00:20:45) Rabbit shows off the AI agent it should have launched with (00:22:23) Mistral’s Le Chat tops 1M downloads in just 14 days Applications & Business (00:24:06) OpenAI Tops 400 Million Users Despite DeepSeek’s Emergence (00:27:37) Google’s new AI video model Veo 2 will cost 50 cents per second (00:29:52) HP is buying Humane and shutting down the AI Pin Projects & Open Source (00:31:44) Microsoft launches next-gen Phi AI models. (00:33:47) OpenAI introduces SWE-Lancer: A Benchmark for Evaluating Model Performance on Real-World Freelance Software Engineering Work (00:37:12) SWE-Bench+: Enhanced Coding Benchmark for LLMs Research & Advancements (00:40:00) Towards an AI co-scientist (00:42:52) Magma: A Foundation Model for Multimodal AI Agents Policy & Safety (00:47:32) Demonstrating specification gaming in reasoning models (00:51:03) Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
--------
58:37
#200 - ChatGPT Roadmap, Musk OpenAI Bid, Model Tampering
Our 200th episode with a summary and discussion of last week's big AI news! Recorded on 02/14/2025 Join our brand new Discord here! https://discord.gg/nTyezGSKwP Hosted by Andrey Kurenkov and Jeremie Harris. Feel free to email us your questions and feedback at [email protected] and/or [email protected] Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. In this episode: OpenAI announces plans to unify their model offerings, moving away from multiple separate models (GPT-4, Claude, etc.) toward a single unified intelligence system, with free users getting "standard intelligence" and Plus subscribers accessing "higher intelligence" levels. Adobe launches their Sora-rivaling AI video generator with 1080p output and 5-second clips, emphasizing production-ready content for films and introducing new pricing tiers through Firefly subscriptions at $10-30 per month. Elon Musk and a consortium offer $97.4 billion to acquire OpenAI's nonprofit entity, potentially complicating the company's transition to a for-profit structure, though Sam Altman quickly dismissed the offer's viability. TSMC implements stricter chip sales restrictions to China, requiring government-approved third-party packaging houses for chips using 16nm and below processes, aligning with US export control measures and affecting major tech companies like Nvidia and AMD. Timestamps + Links: (00:00:00) Intro / Banter (00:01:25) Response to listener comments (00:02:41) News Preview Tools & Apps (00:03:58) Adobe’s Sora rivalling AI video generator is now available for everyone (00:09:45) OpenAI lays out plans for GPT-5 (00:16:42) OpenAI is rethinking how AI models handle controversial topics (00:21:28) Perplexity AI launches new ultra-fast AI search model Sonar (00:23:45) YouTube AI updates include auto dubbing expansion, age ID tech, and more Applications & Business (00:24:37) Musk-led group makes $97.4 billion bid for control of OpenAI (00:34:32) Anthropic’s next major AI model could arrive within weeks (00:39:09) AI chip startup Groq secures $1.5 billion commitment from Saudi Arabia (00:42:15) OpenAI reportedly planning to build its first AI chip in 2026 Projects & Open Source (00:45:01) Zyphra Introduces the Beta Release of Zonos: A Highly Expressive TTS Model with High Fidelity Voice Cloning (00:51:11) Gemstones: A Model Suite for Multi-Faceted Scaling Laws (00:57:15) Hephaestus: Improving Fundamental Agent Capabilities of Large Language Models through Continual Pre-Training Research & Advancements (00:58:24) Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities (01:04:24) Distillation Scaling Laws (01:10:06) Matryoshka Quantization (01:17:47) How much AI compute exists globally? How rapidly is it growing? Policy & Safety (01:21:29) US and UK refuse to sign summit declaration on AI safety (01:25:43) Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs (01:34:40) xAI Risk Management Framework (Draft) (01:39:59) TSMC bans more chip sales to China due to stricter U.S. export sanctions (01:42:38) Listener requested topic Synthetic Media & Art (01:43:48) Thomson Reuters Wins First Major AI Copyright Case in the US (01:44:46) Scarlett Johansson calls for deepfake ban after AI video goes viral (01:45:55) Outro
--------
1:48:09
#199 - OpenAI's 03-mini, Gemini Thinking, Deep Research, s1
Our 199th episode with a summary and discussion of last week's big AI news! Recorded on 02/09/2025 Join our brand new Discord here! https://discord.gg/nTyezGSKwP Hosted by Andrey Kurenkov and Jeremie Harris. Feel free to email us your questions and feedback at [email protected] and/or [email protected] Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. In this episode: - OpenAI's deep research feature capability launched, allowing models to generate detailed reports after prolonged inference periods, competing directly with Google's Gemini 2.0 reasoning models. - France and UAE jointly announce plans to build a massive AI data center in France, aiming to become a competitive player within the AI infrastructure landscape. - Mistral introduces a mobile app, broadening its consumer AI lineup amidst market skepticism about its ability to compete against larger firms like OpenAI and Google. - Anthropic unveils 'Constitutional Classifiers,' a method showing strong defenses against universal jailbreaks; they also launched a $20K challenge to find weaknesses. Timestamps + Links: (00:00:00) Intro / Banter (00:02:27) News Preview (00:03:28) Response to listener comments Tools & Apps (00:08:01) OpenAI now reveals more of its o3-mini model’s thought process (00:16:03) Google’s Gemini app adds access to ‘thinking’ AI models (00:21:04) OpenAI Unveils A.I. Tool That Can Do Research Online (00:31:09) Mistral releases its AI assistant on iOS and Android (00:36:17) AI music startup Riffusion launches its service in public beta (00:39:11) Pikadditions by Pika Labs lets users seamlessly insert objects into videos Applications & Business (00:41:19) Softbank set to invest $40 billion in OpenAI at $260 billion valuation, sources say (00:47:36) UAE to invest billions in France AI data centre (00:50:34) Report: Ilya Sutskever’s startup in talks to fundraise at roughly $20B valuation (00:52:03) ASML to Ship First Second-Gen High-NA EUV Machine in the Coming Months, Aiming for 2026 Production (00:54:38) NVIDIA’s GB200 NVL 72 Shipments Not Under Threat From DeepSeek As Hyperscalers Maintain CapEx; Meanwhile, Trump Tariffs Play Havoc With TSMC’s Pricing Strategy Projects & Open Source (00:56:49) The Allen Institute for AI (AI2) Releases Tülu 3 405B: Scaling Open-Weight... (01:00:06) SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model (01:03:56) PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models (01:08:26) OpenEuroLLM: Europe’s New Initiative for Open-Source AI Development Research & Advancements (01:10:34) LIMO: Less is More for Reasoning (01:16:39) s1: Simple test-time scaling (01:19:17) ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning (01:23:55) Streaming DiLoCo with overlapping communication: Towards a Distributed Free Lunch Policy & Safety (01:26:50) US sets AI safety aside in favor of 'AI dominance' (01:29:39) Almost Surely Safe Alignment of Large Language Models at Inference-Time (01:32:02) Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming (01:33:16) Anthropic offers $20,000 to whoever can jailbreak its new AI safety system
--------
1:37:46