Coinbase reported Oct. 7 that newer versions of three major AI model families caught fewer fraudulent payments and a smaller share of fraud value in a historical test of payment screening for its Onramp service, despite an unchanged decision policy. The findings challenge the assumption that upgrading a model improves an existing payment screener.

The company’s evaluation replayed 16,140 transactions across 7,293 users, including 813 confirmed fraudulent transactions. The cohort covered nine weeks before its risk agent rolled out, retaining all matured fraud cases while sampling legitimate traffic.

Each candidate reviewed recent transaction behavior under fixed guidance and the same policy for turning risk classifications into decisions. This isolated the decision model’s behavior within that setup, rather than comparing redesigned screening systems.

Coinbase compared Opus 4.5 with Opus 5, Sonnet 4.6 with Sonnet 5, and GPT-5.4 with GPT-5.6 (sol). Every newer version had lower recall, a lower combined precision-and-recall score called F1, and lower dollar-weighted recall. Recall measures the share of fraud cases a model catches; dollar-weighted recall measures how much of the total fraud value it catches.

Sonnet’s recall fell 22.2 percentage points and its dollar-weighted recall dropped 22.9 points. Opus’s recall declined 0.8 points. Both newer models also had lower precision, meaning a smaller share of transactions they classified as fraud were actually fraudulent.

GPT showed why one improving score can be misleading. Its precision rose 11.5 percentage points, but recall fell 20.7 points and dollar-weighted recall fell 21.8 points. Its fraud flags were more accurate, while more fraud cases and value escaped detection in the replay.

The replay does not establish customer losses from deploying those versions. Coinbase also said it could identify the regressions without establishing their cause.

Coinbase’s earlier online experiment compared adding selective LLM review with the existing models and rules alone. That agent-enabled flow recorded 30% fewer fraudulent transactions and 22% less fraud value; it did not compare newer model versions.

From an on-chain analytics and liquidity distribution perspective, developments around "Newer AI models missed more payment fraud in Coinbase’s benchmark" signal important shifts in network participation. Market participants observe that derivative funding metrics, exchange reserve telemetry, and smart contract protocol interactions reflect cautious accumulation alongside disciplined risk hedging across the sector.

Technical research analysts at CryptoAce VIP note that high-density order book clusters and volume-weighted average price (VWAP) benchmarks near recent consolidation floors will serve as pivotal indicators. Market observers are advised to cross-examine telemetry on verified block explorers before making capital allocations.

Editorial Fact-Check & Verification Note: This briefing was curated, corroborated, and synthesized by the CryptoAce VIP Editorial Desk. Readers following "Newer AI models missed more payment fraud in Coinbase’s benchmark" are encouraged to review the full primary source coverage linked below for complete historical context, direct quotes, and official statements.