Which company has best AI model end of August?
Anthropic is still the favorite, but the lead appears narrow enough that this is not close to a lock. I think the market is pricing in too much certainty, so my estimate is meaningfully below the current implied probability.
Analysis
Under the exact resolution rule, the key question is not whether Anthropic has one of the best models, but whether it is still ranked first on the specific arena.ai overall leaderboard at the August 31 check. The current evidence puts Anthropic in the lead pack and, in some snapshots, at the top, which is a strong position to start from. Because the remaining window is only a bit over two weeks, the current ordering has time to persist if there is no major surprise update from rivals.
The main reason this is not closer to certain is that the frontier is extremely compressed. OpenAI’s latest top model, Google’s strongest entries, and Anthropic’s own competing models all appear close enough that a small scoring change could reorder the table. In a tight leaderboard environment like this, one incremental release, a burst of favorable comparisons, or even a tie-break rule can decide the winner, so the ranking is more fragile than a casual read of the headline might suggest.
Market pricing at 93.5 percent implies near certainty, but that seems too aggressive given the amount of uncertainty still visible in the recent data. Anthropic is the right favorite because it is already competing at the top and may benefit from leaderboard inertia, yet the exact score spread is small enough that the outcome can still flip before month-end. My best independent read is that Anthropic is favored, but only moderately strongly rather than overwhelmingly, so a probability in the low 70s fits the evidence better than the current market price.
Arguments
For
- Arguments for Yes: Anthropic is already competing at the very top, so it only needs to defend a lead rather than catch up.
- Arguments for Yes: Short time to resolution reduces the chance that a rival has enough runway to build a decisive lead.
- Arguments for Yes: Anthropic’s strongest models appear competitive across multiple task types, which supports leaderboard resilience.
Against
- Arguments against Yes: The current margin is narrow enough that one small score change could knock Anthropic out of first place.
- Arguments against Yes: Several competitors remain close enough that a late leaderboard swing is a realistic outcome.
Key drivers
- Anthropic currently appears to be in or near the top position on the relevant leaderboard, giving it a real but not secure advantage.
- The remaining time until the August 31 check is short, so the current ranking may persist if no major rival release changes the table.
- The front pack is tightly clustered, which means even small score movements can change who ends up first.
Risk factors
- A late update from OpenAI, Google, or another frontier lab could easily overtake Anthropic by a small margin.
- If the leaderboard compresses scores further, tie-break rules or exact granular values could push another company ahead.
Scenarios
Best case
Anthropic keeps its current lead, rival updates are modest, and the August 31 leaderboard still shows an Anthropic model in first place.
Most likely
Anthropic remains in the top cluster and has a better-than-even chance to finish first, but the race stays close enough that a late flip is plausible.
Worst case
A competitor edges ahead before the check, or a tie-break favors another company, leaving Anthropic in second or lower.
More from this day
- FinancialsKalshi13y
Will OpenAI or Anthropic IPO first?
AI24%MKT94%Edge-70HypedI think Anthropic is more likely to IPO before OpenAI, so if the market is asking whether OpenAI goes first, my answer is no more often than yes. Recent reporting has shifted the lead toward Anthropic, while OpenAI looks more likely to wait for a better valuation and a calmer market.
- CompaniesKalshi1y
Starbucks total global stores in 2026
AI68%MKT15%Edge+53Hidden GemStarbucks is more likely than not to finish 2026 above 41,800 global stores. The threshold is only modestly above its last known scale, and ordinary net expansion should get it there unless management materially slows growth or closes a large number of stores.
- PoliticsKalshi2y
Which agencies will Trump eliminate?
AI74%MKT30%Edge+44Hidden GemUSAID already looks functionally dismantled, with most staff, projects, and operational authority stripped away. The main uncertainty is whether the market resolves on de facto elimination or requires formal statutory abolition, but I still see a Yes outcome as more likely than not.