The release of Google's Gemini 3.5 model series on May 19, 2026, has catalyzed a significant repricing in the market for the "Best AI at the end of 2026," with traders increasing the implied probability of a Google victory [3]. In the session on June 03, 2026, contracts for 'Gemini' to be the top-ranked large language model (LLM) at year-end rose a notable 5.0 percentage points, from 15.3% to 20.3%. This shift came primarily at the expense of the market's frontrunner, Anthropic's 'Claude,' suggesting that traders view the latest Gemini release as a meaningful challenge to the incumbent's long-term dominance.
The repricing reflects a broader sentiment that the competitive landscape in artificial intelligence is tightening. While Anthropic's Claude series has maintained a lead on key benchmarks for much of the year, the gains for Google's Gemini—along with smaller increases for OpenAI's ChatGPT and xAI's Grok—indicate that probability is being redistributed from a single leader to a more competitive field of challengers. The market, which resolves based on rankings from the LM Arena Leaderboard, now implies a more contested race to the end-of-year deadline [6].
Distribution Analysis
| Outcome | Current Prob | Change | Volume |
|---|---|---|---|
| Claude | 60% | -3.8pp | 11,954 |
| Gemini | 18% | +5.0pp | 9,034 |
| ChatGPT | 14% | +1.3pp | 10,415 |
| Grok | 6% | +1.2pp | 6,370 |
| Muse Spark | 2% | ~0pp | 13,596 |
| Qwen | 0% | ~0pp | 134 |
| Ernie | 0% | ~0pp | 134 |
Net: 3 of 7 contracts rose on a combined volume of 25,819, shifting the implied consensus away from the long-time leader 'Claude' and toward its key competitors.
What's Driving the Shift
The move appears to be driven by the market digesting the capabilities of recently announced models, particularly Google's latest offering.
Google's Gemini 3.5 Release: The primary catalyst for the repricing is Google's May 19 announcement of Gemini 3.5 [3]. The release highlighted the model's "frontier performance for agents and coding," noting its ability to outperform previous versions and rival flagship models from competitors on complex, long-horizon tasks. This focus on agentic capabilities, which are seen as crucial for enterprise adoption and real-world utility, appears to have convinced traders to increase their assessment of Google's long-term standing [9].
A Tightening Competitive Field: The market shift coincides with a period of rapid innovation from all major AI labs. OpenAI released GPT-5.5 on April 23, and Anthropic announced Claude Opus 4.8 on May 28 [1], [2]. While these were strong releases, the market's reaction suggests Google's Gemini 3.5 was perceived as a particularly potent competitive development. This aligns with analysis from May 2026 suggesting that the top models from Anthropic, Google, and OpenAI were already in a "statistical tie" on general text tasks, making the leaderboard highly susceptible to reshuffling [5].
Reallocation from the Frontrunner: The 5.0-point gain for 'Gemini' was funded almost entirely by a 3.8-point drop for 'Claude.' This direct reallocation suggests that traders are not simply more bullish on the AI space as a whole, but are specifically re-evaluating the competitive moat of Anthropic's models in light of Google's recent progress.
Market Context
For much of 2026, contracts on the CFTC-regulated Kalshi exchange have priced Anthropic's 'Claude' as the clear favorite to end the year as the top-ranked LLM. This was supported by its leading position on the LM Arena Leaderboard, a key public benchmark that uses crowdsourced human votes to rank models [4], [6]. As of May 2026, Claude Opus 4.6 held the top spot, though its lead over rivals from Google and OpenAI had narrowed significantly [5].
The pace of frontier model releases has accelerated to a point where "any analysis of 'the leaderboard' that is older than two weeks is describing a market that no longer exists," according to an analysis by Swfte AI [4]. This rapid cadence creates a volatile environment for prediction markets, where a single strong release can trigger a substantial re-evaluation of the competitive hierarchy. The recent move in 'Gemini' contracts is the latest evidence of this dynamic, as traders price in a more uncertain and competitive race.
What to Watch
The primary factor for this market is the release schedule and demonstrated performance of new flagship models from the leading AI labs. With the market still more than six months from its December 31, 2026, close, there is ample time for further disruption. Traders will be closely watching for announcements regarding OpenAI's next major release, potentially GPT-6, which other prediction markets are already pricing for a late 2026 debut [10]. The market will ultimately settle based on the model holding the #1 rank on the "Text Arena | Overall" leaderboard at lmarena.ai on the expiration date [7].
