What are the key takeaways from this Machine Learning Street Talk episode?

Key insights include: **Sovereign AI compute strategy:** Cosine AI received government-allocated compute on the ISAMBARD supercomputer cluster in Bristol through the UK's Sovereign AI Unit, eliminating the largest startup barrier to frontier model training. Without this allocation, a comparable project would consume a significant portion of a $50–100M raise, making the effort financially unviable for a company of Cosine's size.; **Inference-free licensing model:** Cosine deploys model weights directly to customer GPUs or customer-managed cloud environments like Azure and AWS, rather than hosting inference themselves. This means revenue comes from technology licensing, not token margins, dramatically reducing infrastructure costs and making frontier model development viable without the billions Anthropic or OpenAI spend on inference data centers.; **Active parameter count over total parameter count:** Open-source models like DeepSeek V3 have 1.6 trillion total parameters but only 30–50 billion active per token, optimized for inferability rather than raw performance. Closed models like Claude Sonnet likely run 100+ billion active parameters from a 1.1–1.5 trillion total MOE architecture, which Pullen argues explains the persistent performance gap between open and closed frontier models.

What did Alistair Pullen discuss on Machine Learning Street Talk?

Alistair Pullen, CEO of Cosine AI, explains how his UK-based frontier lab secured government compute allocation on the ISAMBARD supercomputer to build Britain's first sovereign LLM, why nations cannot rely on US AI access, and how constrained resources force architectural and algorithmic innovation in model training. Key topics include: **Sovereign AI compute strategy:** Cosine AI received government-allocated compute on the ISAMBARD supercomputer cluster in Bristol through the UK's Sovereign AI Unit, eliminating the largest startup barrier to frontier model training. Without this allocation, a comparable project would consume a significant portion of a $50–100M raise, making the effort financially unviable for a company of Cosine's size.; **Inference-free licensing model:** Cosine deploys model weights directly to customer GPUs or customer-managed cloud environments like Azure and AWS, rather than hosting inference themselves. This means revenue comes from technology licensing, not token margins, dramatically reducing infrastructure costs and making frontier model development viable without the billions Anthropic or OpenAI spend on inference data centers..

How long is this episode of Machine Learning Street Talk?

This episode is 55 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

Machine Learning Street Talk

Why a Nation Can't Outsource Its Frontier AI - Alistair Pullen (Cosine AI)

July 13, 2026

55 min episode · 2 min read

Alistair Pullen

Episode

55 min

Read time

2 min

Topics

Productivity, Startups, Fundraising & VC

AI-Generated Summary

Published Jul 14, 2026

Key Takeaways

✓Sovereign AI compute strategy: Cosine AI received government-allocated compute on the ISAMBARD supercomputer cluster in Bristol through the UK's Sovereign AI Unit, eliminating the largest startup barrier to frontier model training. Without this allocation, a comparable project would consume a significant portion of a $50–100M raise, making the effort financially unviable for a company of Cosine's size.
✓Inference-free licensing model: Cosine deploys model weights directly to customer GPUs or customer-managed cloud environments like Azure and AWS, rather than hosting inference themselves. This means revenue comes from technology licensing, not token margins, dramatically reducing infrastructure costs and making frontier model development viable without the billions Anthropic or OpenAI spend on inference data centers.
✓Active parameter count over total parameter count: Open-source models like DeepSeek V3 have 1.6 trillion total parameters but only 30–50 billion active per token, optimized for inferability rather than raw performance. Closed models like Claude Sonnet likely run 100+ billion active parameters from a 1.1–1.5 trillion total MOE architecture, which Pullen argues explains the persistent performance gap between open and closed frontier models.
✓Credit attribution in RL training: Standard RL assigns uniform weight to every token in a 256,000-token rollout based solely on a final pass/fail signal, reinforcing both correct decisions and irrelevant filler equally. Cosine targets high-entropy decision points within trajectories, attributing advantage disproportionately to those tokens, producing more targeted weight updates, better sample efficiency, and models that learn reusable abstractions rather than surface-level patterns.
✓Synthetic grader generation for coding RL: Most real-world software engineering tasks — feature work, refactoring — lack built-in test suites, making RL reward signals impossible without additional infrastructure. Cosine built an 18-month pipeline using custom post-trained models to synthesize implementation-agnostic graders for arbitrary codebases, enabling RL data generation across languages including Java, Fortran, C++, Verilog, and SysVerilog without requiring pre-existing test frameworks.

What It Covers

Alistair Pullen, CEO of Cosine AI, explains how his UK-based frontier lab secured government compute allocation on the ISAMBARD supercomputer to build Britain's first sovereign LLM, why nations cannot rely on US AI access, and how constrained resources force architectural and algorithmic innovation in model training.

Key Questions Answered

•Sovereign AI compute strategy: Cosine AI received government-allocated compute on the ISAMBARD supercomputer cluster in Bristol through the UK's Sovereign AI Unit, eliminating the largest startup barrier to frontier model training. Without this allocation, a comparable project would consume a significant portion of a $50–100M raise, making the effort financially unviable for a company of Cosine's size.
•Inference-free licensing model: Cosine deploys model weights directly to customer GPUs or customer-managed cloud environments like Azure and AWS, rather than hosting inference themselves. This means revenue comes from technology licensing, not token margins, dramatically reducing infrastructure costs and making frontier model development viable without the billions Anthropic or OpenAI spend on inference data centers.
•Active parameter count over total parameter count: Open-source models like DeepSeek V3 have 1.6 trillion total parameters but only 30–50 billion active per token, optimized for inferability rather than raw performance. Closed models like Claude Sonnet likely run 100+ billion active parameters from a 1.1–1.5 trillion total MOE architecture, which Pullen argues explains the persistent performance gap between open and closed frontier models.
•Credit attribution in RL training: Standard RL assigns uniform weight to every token in a 256,000-token rollout based solely on a final pass/fail signal, reinforcing both correct decisions and irrelevant filler equally. Cosine targets high-entropy decision points within trajectories, attributing advantage disproportionately to those tokens, producing more targeted weight updates, better sample efficiency, and models that learn reusable abstractions rather than surface-level patterns.
•Synthetic grader generation for coding RL: Most real-world software engineering tasks — feature work, refactoring — lack built-in test suites, making RL reward signals impossible without additional infrastructure. Cosine built an 18-month pipeline using custom post-trained models to synthesize implementation-agnostic graders for arbitrary codebases, enabling RL data generation across languages including Java, Fortran, C++, Verilog, and SysVerilog without requiring pre-existing test frameworks.

Notable Moment

Pullen describes feeling like a second-class citizen for the first time when US export controls blocked access to frontier models like Fable. Rather than accepting this as permanent, he frames it as the core motivation driving Cosine's sovereign model effort, calling it a situation with no option but to succeed.

Know someone who'd find this useful?

Episode Transcript

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome? That's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration blog, or finally break down that long article you've had open for weeks, Gemini and Chrome is here for it. Ready to make anything online make sense? There's no place like Chrome. Check responses set up required compatibility and availability varies 18 plus. When you need to build up your team to handle the growing chaos at work, use Indeed sponsored jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications, and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a 75 sponsored job credit at indeed.com/podcast. That's indeed.com/podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed sponsored jobs. I was horrified as many were when Fable suddenly, got banned. We have obtained the mandate to build, The UK's first sort of sovereign LLM. How can you do in millions what they are doing with billions? There isn't a huge amount of room for error or a huge amount of wiggle room. It was not something that was on my bingo card in January. The number's like 10 trill being knocked around. We we were compressing most of the Internet at that point. It's really funny. After Claude deletes your production database, it it it'll say, oh, oh, you're right to point that out. Agentic harnesses are getting less important over time. A model can probably do with bash only, basically, any task these days. Also, thank you, Donald Trump. Yeah. I mean, for the first time, I feel like a second class citizen because they are going faster than I am. And I I really hate that. That boils my blood more than anyone else, and I we are going to do everything we can to pull this off. We have no choice but to make it happen. Alastair, it's it's great to meet you, mate. We are here in London, where it's customary to say hello, geyser. Oh, geyser. So where in London are we? We are in Hoxton right now. So we're in Shoreditch. We're about half a mile away from where Cosine started in my apartment, which was in Hoxton Square, just over there. So we haven't come very far, but we have expanded a fair bit since then. And what is Cosine? Cosign is a frontier lab based here in The UK. Prior to about three months ago, we built, best in class coding agents specifically for, highly regulated and high side environments. So think things like financial services, insurance, defense, and so on. But more recently, we have obtained the mandate to build, The UK's first sort of sovereign LLM, which …

Get the full transcript (11,866 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Machine Learning Street Talk transcripts →

You just read a 3-minute summary of a 52-minute episode.

Get Machine Learning Street Talk summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Part of this week's recap (Jul 13 – Jul 19)

Keep Reading

Stealing Reasoning Traces from Proprietary LLM APIs — Ilia Shumailov & Alexander Panfilov

Aug 22 · 49 min

Latent Space

The Professor of Outputmaxxing — Anjney Midha, AMP

Jun 18

Every Exponential Ends — Silicon Valley Forgot — Adam Becker

Aug 20 · 78 min

The AI Breakdown

AI Inequality

May 17

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links. As an Amazon Associate, SignalCast earns from qualifying purchases.

Tools

DeepSeek V3
by DeepSeek
“Open-source models like DeepSeek V3 have 1.6 trillion total parameters but only 30–50 billion active per token”
Claude Sonnet
by Anthropic
“Closed models like Claude Sonnet likely run 100+ billion active parameters from a 1.1–1.5 trillion total MOE architecture”
AWS
by Amazon
“Cosine deploys model weights directly to customer GPUs or customer-managed cloud environments like Azure and AWS”

Gear

ISAMBARD supercomputer
“Cosine AI received government-allocated compute on the ISAMBARD supercomputer cluster in Bristol through the UK's Sovereign AI Unit”
Amazon