How billionaire investors of the internet era are fighting to determine whether AI's future will be one of concentrated safety or unfettered advancement
Alex Konrad / Forbes : X: @bgurley , @sarthakgh , @ylecun , and @twobitidiot X: Bill Gurley / @bgurley : Tickled to be mocked a “johnny-come-lately” to open-source in @alexrkonrad's important article. Luckily for me, the Internet is a nice archive. Enjoy my first blog post on open-source, weeks away from its 25 year anniversary. I think it holds up well. https://abovethecrowd.com/... Sar Haribhakti / @sarthakgh : Great piece by @alexrkonrad on the civil war in AI “Neither side believes the other's future is a realistic outcome; both argue that, should their reasoning prevail, there's no need to make a choice.” https://www.forbes.com/... Yann LeCun / @ylecun : The Big AI Debate explained by Forbes. What is more beneficial or more dangerous: open source AI or proprietary controlled by 3 or 4 big players? The people who worry most about AI safety also tend to be he ones who overestimate the power of AI. Ryan Selkis / @twobitidiot : I am not exaggerating when I call @vkhosla & @reidhoffman fascists: the Silicon Valley tech god enablers that want to capture and dominate AI for their own enrichment. These people (plus Altman) are the most dangerous people in our world today. That will be obvious with time.
Context & Ripple Effects
The argument arrives after safety-focused philanthropy had already sought to shape public-sector attention to long-term AI risks through Open Philanthropy's support for policy staff. That gives the investor dispute relevance beyond rhetoric: it concerns which assumptions reach policymakers and funders.
The open-access side also has institutional stakes. Earlier coverage described an open-source ecosystem dependent on large companies' model releases, while the Hassabis-LeCun exchange made the safety-versus-control disagreement explicit in public.
First-order effects
- The article publicly hardens the split between backers of concentrated safety governance and advocates of more open, fast-moving AI development, giving figures such as Bill Gurley and Yann LeCun clearer positions to defend.
- Safety advocates and open-source proponents gain a fresh focal point for contesting whether restrictions are necessary safeguards or a route to control by dominant players.
Second-order effects
- AI companies, nonprofits and investors face greater pressure to explain how their model-access policies align with either safety controls or broad availability; the debate is especially consequential where open projects rely on Big Tech model releases.
- Policy advocacy may become more contested as each camp tries to frame the other's preferred outcome—regulatory concentration or insufficient safeguards—as the larger risk.
Third-order effects
- If this divide persists, AI governance may increasingly be shaped by a contest over who controls frontier models and access to them, rather than by a shared safety baseline.
- The durability of genuinely open alternatives will remain a structural test: projects such as the nonprofit OLMo effort offer one counterweight, but do not by themselves resolve dependence on well-resourced model providers.
The trend: AI policy is becoming a struggle over whether safety governance concentrates control of advanced models or whether broader access is itself a safeguard against concentrated power.