Scan: Will any company announce that it has achieved Artificial General Intelligence (AGI) before Oct 1, 2026?

date	predicted	market price	confidence
2026-09-03	5%	21%	76%

1. Planner Fable 5.1 · Fermi decomposition + tool plan

Tools chosen: kalshi_direct kalshi_related polymarket_related article_search claude_news gdelt_news wikipedia

Sub-questions (Fermi decomposition)

What is the current Kalshi price and trading history for KXAGICO-COMP-26Q3, and have there been meaningful price moves recently?
Have any AI labs (OpenAI, Google DeepMind, Anthropic, xAI, Meta, DeepSeek, or smaller startups) made statements in the last 90 days suggesting they intend to declare AGI in 2026?
What is the status of the OpenAI–Microsoft agreement regarding an AGI declaration and the expert-panel verification process, and does OpenAI have an incentive to declare AGI before Oct 2026?
What are Polymarket and other Kalshi markets pricing for AGI announcements or related milestones (e.g., 'OpenAI announces AGI', 'AGI by 2026/2027') and do they disagree with this market?
Have any companies (including minor or fringe ones) already publicly claimed to have achieved AGI, and would such claims plausibly count under Kalshi's resolution language?
What major frontier model releases (GPT-6, Gemini 3/4, Claude, Grok 5) are expected before Oct 2026, and have executives tied any of them to AGI claims?

Planner reasoning

This is a Kalshi question about whether any company will officially announce achieving AGI before Oct 1, 2026. Key drivers are the current market price, whether any lab (OpenAI, Google DeepMind, Anthropic, xAI, Meta, or smaller firms) has signaled imminent AGI claims, the OpenAI-Microsoft AGI-declaration contractual dynamics, and how loosely 'announce' could be interpreted for resolution. Cross-market signals from Polymarket AGI markets and news on frontier model releases and executive statements are the main empirical inputs.

2. Research Parallel tool calls

tool	status	time	items	summary
kalshi_direct	OK	0.8s	1	## This Market's Kalshi Data Before Oct 1, 2026 (KXAGICO-COMP-26Q3) - Current price (probability): 7.00% - 7-day price change: -22.00% - 30-day price change: +1.00% - Average daily volume: 552 contracts - Price range: 1.00% - 29.00% - Data points: 35 days
kalshi_related	OK	22.9s	3	3 related markets / summaries. series KXAGICO: 0 markets (skipped 12 no-signal) \| keyword 'AGI': ok \| keyword 'artificial general intelligence': ok \| keyword 'OpenAI AGI': ok
polymarket_related	OK	23.0s	0	Scanned 100 active Polymarket markets, kept 0 matches. keyword 'AGI': 0 markets \| keyword 'artificial general intelligence': 0 markets \| keyword 'OpenAI announce AGI': 0 markets
article_search	OK	23.6s	33	Fetched 33 unique articles across 4 queries (lookback=60d). 'company announces achieved artificial general intelligence AGI': 10 articles \| 'OpenAI AGI declaration Microsoft agreement expert panel': 10 articles \| 'Sam Altman Demis Hassabis AGI timeline 2026 statement': 10 articles \| 'startup claims
claude_news	OK	25.5s	13	## Key Findings (as of Sept 10, 2026) - OpenAI's GPT-6 Astra launch (Sept 3, 2026) — closest thing to an "AGI announcement" so far, but stops short of formal declaration. OpenAI released GPT-6 Astra, which president Greg Brockman called a "generational leap" and said could eventually be seen a
gdelt_news	OK	151.1s	20	GDELT: 20 articles across 3 queries (lookback=60d). 'achieved AGI announcement': 10 hits \| 'artificial general intelligence achieved company claims': 10 hits \| 'OpenAI declares AGI': error GDELT rate-limited after retries (429)
wikipedia	FAILED	23.0s	0	http error:

3. Evidence Brief Sonnet 5 · 7333 chars

# Current state Kalshi YES is priced at 7% (down sharply from a recent peak in the 20-29% range, -22% in the last 7 days) after OpenAI's Sept 3, 2026 GPT-6 Astra launch triggered "AGI era" rhetoric from President Greg Brockman that stopped short of a formal AGI declaration. No company has issued an unambiguous, formal announcement that it has achieved AGI as of the latest research (Sept 10, 2026); market resolution requires an "official" announcement, which has not yet occurred despite heavy hype-adjacent language. # Timeline of key events - 2026-04-27: Microsoft–OpenAI contract's "AGI clause" (which had an independent expert-panel verification mechanism for AGI declarations, effective Oct 2025) was removed/renegotiated, eliminating the structural/contractual trigger for a formal AGI announcement (confirmed — Simon Willison, mindstudio.ai). - 2026-07-22: Amazon confirmed layoffs within its internal "AGI unit," signaling organizational activity around AGI research but no capability claim (confirmed — CNBC). - 2026-08-06: Google DeepMind's Demis Hassabis said AGI is "close at hand" while transitioning to a strategic role, ceding day-to-day leadership to Koray Kavukcuoglu (reported — CNBC). - 2026-08-17/08-28: Sam Altman and Chief Research Officer Mark Chen state OpenAI expects an *internal* AGI-level system by year-end 2026 ("80% of the way"), distinct from a public declaration (reported — Digit.in, Forbes). - 2026-09-03: OpenAI launched GPT-6 Astra; President Greg Brockman said "Welcome to the AGI era" and personally believes OpenAI has reached AGI, but explicitly declined to make a formal declaration, calling AGI a "spiritual concept" (confirmed — Axios, betanews). - 2026-09-07/08: Nvidia CEO Jensen Huang publicly declared "AGI has arrived" on social media, reacting to Astra — third-party commentary, not an OpenAI or Nvidia corporate claim (confirmed — Yahoo Finance, widely syndicated iHeart affiliates). # Event Will any company officially announce it has achieved AGI before Oct 1, 2026 (Kalshi ticker KXAGICO-COMP-26Q3)? # Outcomes to forecast - Yes - No # Kalshi market anchor Current YES price: **7%**. 7-day change: -22 points (sharp drop from ~29% peak, likely post-Astra-launch repricing once no formal declaration materialized). 30-day change: +1 point. Average daily volume: 552 contracts. Price range over 35 days: 1%–29%. This is the primary consensus to beat. # Sub-question answers 1. **Kalshi price/history** — 7% currently; peaked near 29% (likely around Astra's launch hype on Sept 3) then fell 22 points in the past week as it became clear no formal declaration was made. [kalshi_direct] 2. **Recent lab statements on 2026 AGI declaration intent** — OpenAI (Brockman) framed GPT-6 Astra as ushering an "AGI era" but explicitly denied a formal declaration; Altman/Chen target an *internal* AGI system by year-end 2026, not necessarily public. DeepMind's Hassabis says AGI is "close" but made no achievement claim. No comparable statements from xAI, Meta, or DeepSeek. [claude_news, article_search] 3. **Microsoft-OpenAI AGI clause status** — The clause requiring expert-panel verification of an AGI declaration was removed on 2026-04-27; this eliminates OpenAI's structural/contractual incentive to formally declare AGI before Oct 2026. [claude_news, Simon Willison] 4. **Cross-market pricing** — No Polymarket AGI markets found (0 matches). Other Kalshi markets (OpenAI/Anthropic IPO race, US stakes in AI labs) don't directly address AGI declaration odds; a cited April 2026 Kalshi data point suggested 55% odds of OpenAI reaching AGI by 2030, implying markets see near-term (2026) declaration as unlikely. [polymarket_related, claude_news] 5. **Fringe/minor company claims** — None identified; only Nvidia's Jensen Huang (not an AI developer of the model in question) made an "AGI has arrived" claim, which is third-party commentary on a rival's product, not a company self-declaration — unlikely to satisfy resolution criteria. [gdelt_news] 6. **Frontier model releases before Oct 2026** — GPT-6 Astra (OpenAI, Sept 3, 2026, reached "Critical" cyber threshold internally); Anthropic's Claude Fable 5.1/Mythos (no AGI framing); Meta's Muse Glimmer (personal/local AI, no AGI claim); Gemini 3.1 Pro (Feb 2026, no major AGI claim). No executive has issued an unambiguous corporate AGI achievement claim. [article_search, claude_news] # Key facts (high-confidence, factual) 1. [kalshi_direct] Current YES price 7%, down from ~29% peak within past 35 days. 2. [claude_news/Simon Willison] Microsoft-OpenAI AGI clause with independent panel verification was scrapped April 27, 2026. 3. [Axios/betanews] Brockman explicitly denied Astra launch constituted a formal AGI declaration. 4. [Forbes] Altman/Chen target internal AGI system by year-end 2026, not public declaration. 5. [Yahoo Finance] Only Nvidia CEO (third party) has used unqualified "AGI has arrived" language. # Cross-market signals - Kalshi related: OpenAI/Anthropic IPO race market implies Anthropic favored (89%) but unrelated to AGI declaration timing. US stakes-in-labs market at 19% — modest confidence in near-term extraordinary government/AI milestones. - Polymarket: No active AGI-specific markets found — no cross-market comparison available. - Sportsbook implied: N/A. # Analyst opinions and speculation - Morningstar analyst notes Google fell behind OpenAI/Anthropic on coding, a key AGI-race front. [CNBC] - Delos.so and others frame 2026 as "AGI: Myth or Reality" — signaling continued ambiguity/skepticism in expert commentary. [article_search] - Claude_news synthesis: "the industry stopped pricing AGI as an event" over the past 12 months — declarations increasingly viewed as marketing rather than discrete technical milestones. # Directional lean per outcome - **Yes**: OpenAI rhetoric ("AGI era," Brockman's personal belief) and Altman's year-end 2026 internal-system timeline keep some probability alive; further model releases (possible GPT-6.x updates, Gemini/Grok releases) before Oct 1, 2026 could produce more aggressive marketing claims. - **No** (favored): No company has made an unambiguous formal declaration; OpenAI explicitly disclaimed one and removed contractual incentive (Microsoft clause gone); competitors (Anthropic, DeepMind, Meta) are notably restrained; Kalshi price already collapsed from 29% to 7% after the closest-yet event (Astra launch) failed to trigger resolution-worthy language. # Gaps / unknowns - Unclear how Kalshi would rule on ambiguous rhetorical claims (e.g., Brockman's "AGI era," Huang's "AGI has arrived") — resolution likely requires a clear, official company self-declaration, which raises the bar substantially. - No visibility into planned Q4 2026 model releases (e.g., GPT-6.5, Gemini 4) or whether any will include formal AGI claims before Oct 1 cutoff. - Polymarket data absence limits cross-market triangulation. # Calibration anchors - Kalshi current YES price: 7% (down from 29% peak) — primary anchor. - Precedent: repeated near-miss "AGI era" framings (OpenAI Sept 2026) that markets treated as insufficient for resolution, causing sharp price decay post-event. - Cited Kalshi data point: ~55% odds (April 2026) of OpenAI reaching AGI by 2030 — suggests market-wide skepticism of a credible declaration this decade, let alone by Oct 2026.

4. Forecasters Fable 5.1 + GPT-5.6 Sol · in parallel

claude-fable-5-1

62% conf

Yes	7%
No	93%

Kalshi prices YES at 7%, having collapsed from a ~29% peak after OpenAI's Sept 3 GPT-6 Astra launch produced 'AGI era' rhetoric but an explicit refusal by Brockman to make a formal declaration—the market judged the closest-yet event as non-qualifying. The Microsoft–OpenAI AGI clause was removed in April 2026, eliminating the main structural incentive for a formal declaration, and Altman/Chen frame their year-end target as an internal system, not a public announcement. Competitors (Anthropic, DeepMind, Meta) remain restrained, and Jensen Huang's 'AGI has arrived' is third-party commentary. With only ~3 weeks to the Oct 1 cutoff and no known scheduled release likely to carry a formal claim, the residual risk lies mainly in resolution ambiguity or a fringe/marketing self-declaration, so I stay at the market anchor.

gpt-5.6-sol

78% conf

Yes	7%
No	93%

The 7% Kalshi YES price is the strongest anchor, especially after its collapse from roughly 29% when GPT-6 Astra produced AGI-adjacent rhetoric but no formal declaration. OpenAI explicitly declined to declare AGI, its stated internal-system target is after the Oct. 1 cutoff, and removal of the Microsoft–OpenAI AGI clause eliminated a prior incentive for a verifiable announcement. DeepMind and other frontier labs remain restrained, no imminent qualifying release or minor-company claim was identified, and Jensen Huang's third-party comment is unlikely to qualify as a corporate achievement announcement. Ambiguous resolution standards and the possibility of an unexpected marketing claim preserve a small YES chance, but the evidence does not justify moving from the market consensus.

5. Devil's Advocate Sonnet 5 · red-team critique

Both forecasters converge exactly on the Kalshi anchor (7%) with almost no independent adjustment, despite the brief noting the close date is only ~3 weeks away and there's a specific catalyst risk: Altman/Chen's "year-end 2026" internal AGI system target could slip earlier or get marketed prematurely — neither forecast explicitly weighs the probability that "internal" system talk bleeds into a public claim before Oct 1.

Neither forecast grapples with resolution ambiguity risk in a differentiated way — the brief flags "unclear how Kalshi would rule on ambiguous rhetorical claims" as a major gap, yet both treat this as a minor residual rather than exploring scenarios where a Kalshi resolution committee might controversially count Brockman's "AGI era" language or a similar future statement as qualifying, which could push YES higher than 7% is pricing.

Both forecasters ignore the base-rate lesson from the recent 29%→7% collapse: this shows the market has been volatile and prone to overreacting to hype, meaning current 7% could still be too high (if further decay is likely) or could snap back up on a new triggering event (e.g., a Gemini or Grok release with aggressive claims) — the analysis doesn't consider directional momentum/decay trajectory, just anchors statically.

The "Yes" case is underexplored: no mention of xAI (Grok) or DeepSeek's incentives — xAI's Musk has a history of making bold/premature technical claims, and this is a known tail risk the brief doesn't rule out explicitly ("No comparable statements from xAI, Meta, or DeepSeek" — but absence of evidence isn't strong evidence of low risk for a company known for hype-driven announcements).

Both forecasters give nearly identical numbers (0.07/0.93) with different confidence levels (0.62 vs 0.78) but no discussion of what would justify the confidence gap — this inconsistency between the two isn't addressed by either, and neither explains why they didn't shade even slightly below 7% given the negative momentum (-22 points in 7 days, still falling) versus the 30-day change of only +1, suggesting further downside drift is plausible before resolution.

The cross-market signal citing "55% odds of OpenAI reaching AGI by 2030" is used as supportive of "No" but neither forecaster stress-tests whether this figure is stale/unreliable (from April 2026, pre-Astra) and could be inconsistent with post-Astra sentiment shifts, representing a missed calibration check.

6. Aggregator Fable 5.1 · final calibrated forecast

6% (70% confidence)

Yes: 6% No: 94%

Both forecasts and the Kalshi anchor sit at 7%, and the critique does not surface new evidence that shifts the picture: the closest-yet catalyst (GPT-6 Astra) produced an explicit refusal by Brockman to formally declare AGI, the Microsoft-OpenAI AGI clause that once incentivized a verifiable declaration is gone, and OpenAI's stated target is an internal system by year-end, after the Oct 1 cutoff. With roughly three weeks remaining and no identified scheduled release from xAI, DeepMind, Meta, or a fringe company carrying AGI-achievement framing, the residual YES probability rests mainly on resolution ambiguity or an unexpected marketing stunt. I shade marginally below the anchor given continued negative price momentum and the short window, but keep meaningful tail mass for a surprise self-declaration or a lenient resolution reading of rhetorical claims.

Pipeline Timing

Total pipeline time: 256.1s

Per-tool research timings shown in the Research section above.