#335 Sriram Raghavan: Why IBM Is Betting Everything on Small AI Models

Why IBM Is Betting Everything on Small AI Models

In this episode of Eye on AI, Craig Smith sits down with Sriram Raghavan, Vice President of AI at IBM Research, to explore one of the most important debates in enterprise AI right now. Do you actually need a massive model to get world class results? IBM's answer is no, and Sriram breaks down exactly why.

Sriram explains why IBM chose to train its Granite models directly using reinforcement learning rather than distilling from larger models like most of the industry. The reason goes beyond performance. It comes down to data lineage, safety alignment, and a belief that small, efficient models are the only sustainable path for enterprises running AI across hybrid cloud environments.

We get into the full technical stack behind that bet. How data quality has replaced model size as the real competitive advantage. Why parameter count is becoming the wrong metric entirely. How IBM's inference time scaling techniques allow an 8 billion parameter model to match the performance of GPT-4o and Claude 3.5 on code and math benchmarks. And why IBM is pioneering a new concept called Generative Computing, which treats AI models not as prompt receivers but as programmable computing elements with runtimes, modular LoRA adapters, and proper programming abstractions.

Sriram also shares where IBM Research is headed next, including breakthroughs in continuous learning, agent orchestration, and making unstructured enterprise data actually usable at scale.

Subscribe for more conversations with the people building the future of AI and emerging technology.

Stay Updated:

Craig Smith on X: https://x.com/craigss

Eye on A.I. on X: https://x.com/EyeOn_AI

(00:00) Why IBM Skips Distillation and Trains Small Models Directly

(04:50) Did We Even Need Giant AI Models in the First Place?

(08:12) How Data Quality Became the New Competitive Moat

(11:54) Why Parameter Count Is the Wrong Way to Measure a Model

(15:36) Reinforcement Learning Without Losing Broad Capabilities

(22:05) Inference Time Scaling: Getting Big Model Results From Small Models

(28:12) Generative Computing: Treating AI as a Programming Element

(36:40) Why IBM Open Sources and How Small Models Make It Sustainable

(41:25) The Path to Continuous Learning Without Rewriting Weights

(51:00) IBM's Full Roadmap: Models, Data, and Agents

Det här avsnittet är hämtat från ett öppet RSS-flöde och publiceras inte av Podme. Det kan innehålla reklam.

Avsnitt(351)

Why the Future of AI Isn't Just Bigger Models. It's Models That Evolve | Risto Miikkulainen of Cognizant

Most AI systems follow a gradient, a mathematical slope that tells them exactly how to improve, step by step, toward a known goal. Neuroevolution doesn't follow any gradient. Instead, it runs hundreds...

2 Juni 1h 4min

How AI Is Reinventing Elder Care | Chia-Lin Simmons of LogicMark

One in four people over 65 will experience a fall, and for most of them, the technology designed to help is a device that hasn't meaningfully changed since the 1980s. Chia-Lin Simmons, CEO of LogicMar...

1 Juni 53min

The App of the Future Is Voice — Not a Screen. Mitel's CTO Luiz Domingos Explains Why.

Luiz Domingos has spent 25 years watching enterprise communications evolve, from IP telephony to cloud to AI, and his assessment of where things stand now is unusually concrete. Companies have moved p...

28 Maj 54min

Is ChatGPT Conscious? A Pioneer of AI Explains | Dr. Terry Sejnowski

A fly with 100,000 neurons can fly, find food, and reproduce. A $100 million supercomputer cannot. Dr. Terry Sejnowski used that observation to silence a room full of MIT AI researchers in the 1980s, ...

28 Maj 56min

Your Child's Data Profile Starts Before They're Born | Eamonn Maguire of Proton

Your child's data profile doesn't start when they get their first phone. It starts before they're born, the moment a parent emails a gynecologist or visits a fertility clinic website. That's the core ...

28 Maj 55min

Training AI Models Without a Billion-Dollar Data Center | Steffen Cruz of Macrocosmos

Training a frontier AI model today requires hundreds of thousands of GPUs, months of compute time, and a budget that only a handful of companies on earth can afford. Steffen Cruz, co-founder and CTO o...

25 Maj 47min

The Single Biggest Barrier to AI Adoption Isn't the Technology — It's This | Errol Gardner of EY

Errol Gardner has spent 35 years advising the world's largest organizations through major technology transitions, and his assessment of where enterprise agentic AI actually stands is one of the most g...

22 Maj 54min

Oliver Dial of IBM: Quantum Advantage Is Happening This Year

IBM's VP of Quantum Systems, Oliver Dial, has spent his career building quantum computers from the ground up, and he's unusually direct about what they can and can't do. In this conversation with Crai...

19 Maj 50min