The MAD Podcast with Matt Turck

What happens when you try to build the “General Electric of AI” with just 14 people? In this episode, Jeremy Howard reveals the radical inside story of Answer AI — a new kind of AI R&D lab that’s not chasing AGI, but instead aims to ship thousands of real-world products, all while staying tiny, open, and mission-driven.

Jeremy shares how open-source models like DeepSeek and Qwen are quietly outpacing closed-source giants, why the best new AI is coming out of China. You’ll hear the surprising truth about the so-called “DeepSeek moment,” why efficiency and cost are the real battlegrounds in AI, and how Answer AI’s “dialogue engineering” approach is already changing lives—sometimes literally.

We go deep on the tools and systems powering Answer AI’s insane product velocity, including Solve It (the platform that’s helped users land jobs and launch startups), Shell Sage (AI in your terminal), and Fast HTML (a new way to build web apps in pure Python). Jeremy also opens up about his unconventional path from philosophy major and computer game enthusiast to world-class AI scientist, and why he believes the future belongs to small, nimble teams who build for societal benefit, not just profit.

Fast.ai

Website - https://www.fast.ai

X/Twitter - https://twitter.com/fastdotai

Answer.ai

Website - https://www.answer.ai/

X/Twitter - https://x.com/answerdotai

Jeremy Howard

LinkedIn - https://linkedin.com/in/howardjeremy

X/Twitter - https://x.com/jeremyphoward

FIRSTMARK

Website - https://firstmark.com

X/Twitter - https://twitter.com/FirstMarkCap

Matt Turck (Managing Director)

LinkedIn - https://www.linkedin.com/in/turck/

X/Twitter - https://twitter.com/mattturck

(00:00) Intro

(01:39) Highlights and takeaways from ICLR Singapore

(02:39) Current state of open-source AI

(03:45) Thoughts on Microsoft Phi and open source moves

(05:41) Responding to OpenAI’s open source announcements

(06:29) The real impact of the Deepseek ‘moment’

(09:02) Progress and promise in test-time compute

(10:53) Where we really stand on AGI and ASI

(15:05) Jeremy’s journey from philosophy to AI

(20:07) Becoming a Kaggle champion and starting Fast.ai

(23:04) Answer.ai mission and unique vision

(28:15) Answer.ai’s business model and early monetization

(29:33) How a small team at Answer.ai ships so fast

(30:25) Why Devin AI agent isn't that great

(33:10) The future of autonomous agents in AI development

(34:43) Dialogue Engineering and Solve It

(43:54) How Answer.ai decides which projects to build

(49:47) Future of Answer.ai: staying small while scaling impact

In this fascinating episode, we dive deep into the race towards true AI intelligence, AGI benchmarks, test-time adaptation, and program synthesis with star AI researcher (and philosopher) Francois Chollet, creator of Keras and the ARC AGI benchmark, and Mike Knoop, co-founder of Zapier and now co-founder with Francois of both the ARC Prize and the research lab Ndea. With the launch of ARC Prize 2025 and ARC-AGI 2, they explain why existing LLMs fall short on true intelligence tests, how new models like O3 mark a step change in capabilities, and what it will really take to reach AGI.

We cover everything from the technical evolution of ARC 1 to ARC 2, the shift toward test-time reasoning, and the role of program synthesis as a foundation for more general intelligence. The conversation also explores the philosophical underpinnings of intelligence, the structure of the ARC Prize, and the motivation behind launching Ndea — a ew AGI research lab that aims to build a "factory for rapid scientific advancement." Whether you're deep in the AI research trenches or just fascinated by where this is all headed, this episode offers clarity and inspiration.

Ndea

Website - https://ndea.com

X/Twitter - https://x.com/ndea

ARC Prize

Website - https://arcprize.org

X/Twitter - https://x.com/arcprize

François Chollet

LinkedIn - https://www.linkedin.com/in/fchollet

X/Twitter - https://x.com/fchollet

Mike Knoop

X/Twitter - https://x.com/mikeknoop

FIRSTMARK

Website - https://firstmark.com

X/Twitter - https://twitter.com/FirstMarkCap

Matt Turck (Managing Director)

LinkedIn - https://www.linkedin.com/in/turck/

X/Twitter - https://twitter.com/mattturck

(00:00) Intro

(01:05) Introduction to ARC Prize 2025 and ARC-AGI 2

(02:07) What is ARC and how it differs from other AI benchmarks

(02:54) Why current models struggle with fluid intelligence

(03:52) Shift from static LLMs to test-time adaptation

(04:19) What ARC measures vs. traditional benchmarks

(07:52) Limitations of brute-force scaling in LLMs

(13:31) Defining intelligence: adaptation and efficiency

(16:19) How O3 achieved a massive leap in ARC performance

(20:35) Speculation on O3's architecture and test-time search

(22:48) Program synthesis: what it is and why it matters

(28:28) Combining LLMs with search and synthesis techniques

(34:57) The ARC Prize structure: efficiency track, private vs. public

(42:03) Open source as a requirement for progress

(44:59) What's new in ARC-AGI 2 and human benchmark testing

(48:14) Capabilities ARC-AGI 2 is designed to test

(49:21) When will ARC-AGI 2 be saturated? AGI timelines

(52:25) Founding of NDEA and why now

(54:19) Vision beyond AGI: a factory for scientific advancement

(56:40) What NDEA is building and why it's different from LLM labs

(58:32) Hiring and remote-first culture at NDEA

(59:52) Closing thoughts and the future of AI research

In this episode, we sat down with Emi Gal, founder and CEO of Ezra, a startup that leverages AI to detect cancer early and inexpensively. Emi provides insights into the landscape of the healthcare sector and talks about the differences between building an AI startup in healthcare versus SaaS. Turns out that "(In AI skills)... are not that transferable." EZRA Website - https://ezra.com Twitter - https://twitter.com/ezrainc Emi Gal LinkedIn - https://www.linkedin.com/in/emigal Twitter - https://twitter.com/emigal FIRSTMARK Website - https://firstmark.com Twitter - https://twitter.com/FirstMarkCap Matt Turck (Managing Director) LinkedIn - https://www.linkedin.com/in/turck/ Twitter - https://twitter.com/mattturck (00:00) Intro (01:50) Ezra raised $21 million in series B round (02:55) The origin of Ezra (06:06) Sourcing AI talent (06:52) Building a proof of concept (09:05) The tipping point for the product market fit (10:57) Y Combinator wants more MRI startups. Why? (11:37) Ezra's vision for MRI (13:25) Is it covered by insurance? (16:15) Full stack vs Software only (20:00) Training AI (22:55) Building an MRI database (25:45) Will radiologists get replaced by AI? (27:52) Creating reports with Generative AI (30:50) Can we trust AI in healthcare? (33:44) What are the specific challenges of building an AI startup? (39:01) Healthcare entrepreneurship (43:59) Staying fit as a CEO: Emi's mental and physical health routine (48:28) Plans for 2024

In this episode, we sat down with Des Traynor, co-founder of Intercom, to explore the seismic shift towards Artificial Intelligence in customer service software. Intercom has gone all-in to embrace AI as people's expectations of what chatbots can do started growing with the release of ChatGPT. Des shares the pivotal moments and strategic decisions that led to this transition, highlighting the urgency and vision that propelled Intercom to integrate AI into their core offerings. Des also delves into the challenges of building a bicontinental startup and the strategic pivot towards becoming an AI-first company. Tune in for an enlightening discussion on the strategy and journey of adapting AI. INTERCOM Website - https://www.intercom.com Twitter - https://twitter.com/intercom Des Traynor LinkedIn - https://www.linkedin.com/in/destraynor/ Twitter - https://twitter.com/destraynor FIRSTMARK Website - https://firstmark.com Twitter - https://twitter.com/FirstMarkCap Matt Turck (Managing Director) LinkedIn - https://www.linkedin.com/in/turck/ Twitter - https://twitter.com/mattturck (00:00) Intro

(01:16) How did Intercom make a transition to a generative AI product (Fin)?

(05:34) Did the Intercom manifesto play a role in the transition?

(07:16) What was the Intercom before Fin?

(09:01) How much development effort did you spend on AI?

(12:31) UX

(15:20) People used to hate chatbots

(17:51) GPT and building layers around it

(20:50) The future of customer service

(23:57) GPT-4/Llama/Mistral/Claude

(25:58) Are multimodal AI-bots the future?

(27:08) AI-hallucination

(30:11) Customization

(34:34) Will Fin get a voice?

(36:26) Customer support cost and impact on profitability

(39:58) How much should you charge?

(45:26) AI-bot resolution rate

(46:43) Can bots take action?

(48:40) AI-adoption

(51:14) How the Intercom team evolve

(53:38) How did 4 Irish guys create a bi-continental startup?

(56:17) Work distribution

(58:38) Tech in Europe vs tech in the US

Today, we’re excited to chat with Scott Belsky - author, entrepreneur, investor and Chief Strategy Officer at Adobe. Matt + Scott discuss the impact of AI on creative work, how Adobe is incorporating AI across their products, and what the future creative tools landscape might look like.

This session was recorded live at a recent Data Driven NYC, our in-person, monthly event series. If you are ever in New York, you can find us on Eventbrite by searching for "FirstMark Capital". Events run monthly and are free and open to everyone. And as always, if you enjoy the MAD podcast, please subscribe and leave us a comment.

Data Driven NYC YouTube Channel

FirstMark Capital Eventbrite

twitter.com/scottbelsky

Implications, by Scott Belsky

twitter.com/mattturck

linktr.ee/mattturck

Show Notes:

[00:53] How Adobe uses AI to enhance user experience, streamline onboarding and automate tasks across their product suite;

[01:30] How AI impacts Adobe's business, making creative processes accessible with features like the context bar in Photoshop;

[02:13] Firefly's journey: internal decisions, training challenges, and a commitment to using licensed material for ethical AI;

[03:58] Moral considerations in Firefly's development: the decision to use licensed material, commercial viability, and addressing user comparisons;

[05:52] Adobe's homegrown approach to generative AI models: in-house development and partnerships for specific capabilities like LLM;

[06:08] Adobe Sensei's 10-year evolution: developing AI technologies, the non-profit Content Authenticity Initiative, and content credentials establishing asset provenance;

[09:17] Adobe's new AI advancements: Firefly Image Model 2, Generative Match, and the vector model for illustration;

[11:16] Firefly Editor's revolutionary image editing: dynamically generating pixels, real-time object manipulation, and Adobe's commitment to pushing technological boundaries;

[12:41] Rapid integration of AI features: Firefly models and playground, surfacing on a website for user testing, and collaboration within Adobe's design organization;

[14:32] How Adobe's AI and data teams are structured and leveraging in-house development for competitive advantage;

[15:47] Future of work and creativity: AI's impact on raising the bar for digital experiences, accelerating creative processes, and the evolving landscape of personalized social content;

[19:11] Leveraging technology to reduce friction, streamline processes, and unlock creative flow;

[20:09] Impact of AI on business models: questioning time-based pricing, anticipating a shift to value-based models, and reconsidering compensation for creative professionals;

[21:10] Parallels with historical Internet Service Providers, the rapid evolution of ideas, and reflections on sustainable business models;

[24:53] Scott’s criteria for evaluating AI investments: valuing skeptical entrepreneurs, acknowledging temporary uniqueness, and emphasizing empathy with customers;

[26:40] Navigating challenges in 2023: Tough decisions for entrepreneurs, evaluating conviction, and the importance of sticking together through the "messy middle”;