Prediction Open AI assessment confidence: 95% certainty 3/5 debate potential 4/5

Jain: Open source models will dominate AI inferencing within two years

Arvind Jain · SemiAnalysis, Altimeter, Nebius, Glean.. 12 Hot Takes From Biggest Names in AI · Sourcery with Molly O'Shea · Jul 19, 2026 · at 31:47

Arvind Jain, founder and CEO of enterprise AI search platform Glean, shares his hottest AI take at the RAISE Summit in Paris.

0:00 / 0:10exact quote · 10.9s

▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark

“Well, I think open source models are going to dominate AI inferencing. You know, you'll see in the next two years, we'll shift to where's almost no open source to where's going to be almost all open source.”

quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →

The claim, plainly

Open source models will dominate AI inferencing within the next two years, shifting from almost no open source usage to almost entirely open source.

a quote alone can be ambiguous, so this restates it with the references and context of the conversation around it resolved. AI-written; the tape below the player is the source.

Assessment

On July 19, 2026, Arvind Jain predicted on the Sourcery podcast that open-source models would dominate AI inferencing 'within the next two years' (establishing a stated horizon of July 19, 2028). As of September 5, 2026, aggregate market telemetry shows open-weight models accounting for approximately 15% to 33% of inference volume (e.g., Presenc AI reporting ~15% in January 2026 and OpenRouter reporting ~30% in late 2025 / ~33% by year-end 2025), with proprietary closed models continuing to process the dominant majority (~70% to 80%) of tokens. Because only seven weeks have elapsed in the stated two-year forecast period and the transition is actively ongoing, the prediction cannot yet be definitively resolved.

MIT Initiative on the Digital Economy / Georgia Tech study (January 20, 2026) found closed models accounted for ~80% of processed tokens on OpenRouter, with open models at ~20%.
OpenRouter's 100-Trillion Token Study (early 2026) showed proprietary models retained ~70% of weekly token volume against ~30% for open-source models by late 2025.
Presenc AI Inference Market Share Report (May 15, 2026) estimated open-weight models accounted for ~15% of aggregate inference market share in January 2026.
The stated two-year evaluation horizon extends through July 19, 2028, leaving nearly the entire prediction window open.

Checked against

[1] leoniemonigatti.com [2] leoniemonigatti.com [3] stanford.edu [4] mit.edu [5] openrouter.ai [6] presenc.ai [7] presenc.ai [8] presenc.ai

open source AI inference volume market share Artificial Analysis open source models share of AI inference volume 2024 2025 open source vs closed source LLM market share inference tokens enterprise LLM open source vs proprietary market share survey

assessed as of 2026-09-05

This assessment is an AI reading of a public statement, made against the cited references with the confidence shown. It can be wrong. The quote, the tape and the references are all on this page; if something is off, use report an error below. full disclaimer →

⏳ prediction date: 2028-07-19; checked again automatically on the next resolution wave

Mentioned

AI inferencing Arvind Jain Glean open source models

⚑ Spot an error? Report it

Jain: Open source models will dominate AI inferencing within two years

Turn any episode into a week of clips.