/Tech4h ago

Nathan Lambert, AI2 post-training lead, warns that upcoming advanced AI models could face bans because they lack political champions

Critics argue national security concerns make tighter regulation inevitable.

1010471211.4K

#80

Original post

Nathan Lambert@natolambert#80inTech

Recent events are so heavy bc that this feels like a start of a new tumultuous era rather than a one & done policy calibration.

It's clearer we need an open ecosystem, but powerful models are coming that could cause strong reactions (or bans) with no champion to defend them.

Interconnects@interconnectsai

Welcome to the AGI era of AI governance It's a one-way door and we weren't ready for it. https://www.interconnects.ai/p/welcome-to-the-agi-era-of-ai-governance

10:55 AM · Jun 14, 2026 · 10.6K Views

Sentiment

Users voiced pessimism about AI governance entering the AGI era because centralized control risks sudden loss of access while open models become politically impossible to defend.

Pos

0.0%

Neg

100.0%

3 comments with sentiment.

Cluster Engagement

Posts from X

Most Activity

VIEWS634LIKES6

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex

@natolambert You're not taking this well I think it was all priced in it was impossible that, so long as DL works, we won't get into the regime where national security concerns start to dominate

Nathan Lambert@natolambert

Recent events are so heavy bc that this feels like a start of a new tumultuous era rather than a one & done policy calibration.

It's clearer we need an open ecosystem, but powerful models are coming that could cause strong reactions (or bans) with no champion to defend them.

3h63460

Nyanpasu@NyanpasuKA

@natolambert Same feeling may even be the start of a new relationship dynamic between humans and AI

4h631

Mehdi Ataei@AtaeiMe

@natolambert All open source models more than certain parameter count will be banned within 6 months.

4h62

Pooja Algikar@algikarpooja

@natolambert yes. the uncomfortable part is that open models may become most necessary exactly when they become most politically difficult to defend. every powerful tool = institutional anxiety test.

3h161

Potato Terminator@econ_tech_vance

@natolambert Feels like the predictable outcome of centralized AI. One call and access is gone.

Mayz@lunan_ai

@natolambert open ecosystem was always the safer bet, and thats exactly why it never wins in timelines like this.

whose stepping up to defend it?