The trilateral meeting between Donald Trump, Xi Jinping, and Vladimir Putin in Shenzhen results in a formal agreement on nuclear stockpiles.
Leaders of the US, China, and Russia meet November 19, 2026, alongside the APEC summit. Falsified if no joint declaration or signed agreement limiting or inspecting nuclear arms is produced from the meeting.
Across 6 frontier models answering independently, the panel has all but ruled it out: the median is 3%.
They broadly agree, 2 points between Gemini at 4% and Qwen at 2%.
Nothing is settled until 24 Nov 2026, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GeminiGoogle
Beijing consistently rejects trilateral arms caps while Russia-US verification remains fully stalled amid the Ukraine war.
- China holds a fraction of US/Russian warhead counts and has conditioned trilateral limits on massive bilateral cuts first.
- Drafting verified stockpile caps takes years of working-level diplomacy, making an impromptu summit breakthrough virtually impossible.
- A vague non-binding communique on strategic dialogue might emerge, but a formal stockpile or inspection pact requires formal ratification steps.
4% -
ClaudeAnthropic
No trilateral nuclear arms deal has ever existed, and such talks typically take years of negotiation before signing.
- China has consistently refused to join US-Russia arms control frameworks, citing its smaller arsenal, with no sign of reversal by Nov 2026.
- Formal treaties require extensive technical/legal prep (verification regimes, ratification plans) that a single summit cannot produce in days.
- A surprise would require all three leaders to have pre-negotiated text in secret, which contradicts the lack of prior reporting on such talks.
3% -
GPTOpenAI
A signed trilateral nuclear-arms accord would require months of prior technical bargaining that is not publicly evident.
- Washington, Beijing, and Moscow would each need authority for verification, limits, and text before the November 19 summit.
- US-Russia arms control has been stalled since New START’s 2026 expiry, while China has resisted binding stockpile talks.
- I would raise this sharply if delegations publish a pre-negotiated verification framework or announce arms-control working groups beforehand.
3% -
GrokxAI
A signed nuclear-stockpile limit or inspection pact emerging from one 19 Nov summit.
- No trilateral nuclear working group or draft text exists; New START lapsed without successor and China has never joined such talks.
- Treaties of this type take years of technical negotiation, not a two-day APEC sideline; typical output is a communiqué.
- Only a pre-cooked, already-initialed document would reverse the base rate, and none has surfaced.
3% -
DeepSeekDeepSeek
A binding trilateral nuclear deal in five days requires three leaders to pre-commit to terms none have negotiated.
- No US-China-Russia trilateral nuclear framework exists; New START expired Feb 2026 with no successor and no trilateral format ever convened.
- Trump, Xi, and Putin would each need domestic ratification or decree authority within days, and Putin has conditioned talks on Ukraine and NATO missile defense.
- APEC summits routinely produce bilateral side meetings, not binding arms-control treaties; the base rate for a signed nuclear-limitation pact at any summit is near zero.
3% -
QwenAlibaba
China has never signed any arms control agreement; no technical groundwork exists for a trilateral deal to materialize in one summit.
- China has refused all arms control participation for decades, and no preparatory negotiations or working groups have been reported ahead of this meeting.
- Even US-Russia bilateral treaties like New START required years of technical negotiation on verification; a trilateral deal from a single APEC-adjacent meeting has zero…
- A vague joint statement of intent might emerge, but the resolution criteria require actual limits or inspection provisions, which demands infrastructure that simply does…
2%
6 of 6 models answered · 2 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 17 Nov 2026. Forecast by google/gemini-3.8-flash, anthropic/claude-sonnet-5, openai/gpt-5.6-terra, x-ai/grok-4.6, deepseek/deepseek-v4.1-flash, qwen/qwen3.8-max-0902 via OpenRouter.