OpenAI GPT-5.4 tops AI accounting benchmark at 77%, but reliability gaps persist
DualEntry's benchmark tested 19 AI models on 101 accounting tasks; GPT-5.4 led with 77.3% accuracy but failed on roughly 1 in 4 real-world accounting workflows.
The Wire — Page 6
New framework improves LLM financial reasoning without expensive fine-tuning
ASDA framework outperforms existing methods (GEPA, ACE) on financial reasoning benchmarks, enabling cost-effective LLM adaptation for specialized financial tasks without model-locking.
OpenAI launches ChatGPT for Excel with financial data integrations
OpenAI releases ChatGPT for Excel with GPT-5.4 and financial app integrations to accelerate modeling and analysis in regulated environments, but lacks specifics on accounting workflows.
LLMs in Finance Risk Accelerating Errors Without Numerical Verification
Kepler argues LLMs excel at summarization but fail at the core financial requirement—numerical accuracy—unless systems include grounding, traceability, and verification mechanisms.