top of page


The AI Trial Balance Trial: low effort, refusals and hallucinations
The AI Trial Balance Trial: Chat GPT low effort failure, Gemini won't even try and Claude tried hard but failed harder. The results were clear: general-purpose AI still can’t handle even basic accounting reporting reliably.
Niv Nissenson
Nov 133 min read


Vibe Excelling: Testing Elkar’s AI Add-On for Excel
Elkar, a new AI add-on for Excel, exceeded my expectations where other AI tools fell short. From handling consolidations correctly to reformatting a 200+ row P&L, it showed real accounting nuance. Elkar proves how AI can save finance teams time inside their most familiar tool.
Niv Nissenson
Aug 203 min read


Polymarket guide to AI models
With so much hype in the AI world, prediction markets like Polymarket offer a refreshing way to cut through the noise. Instead of opinions, people put real money behind their beliefs — giving us a more grounded snapshot of which AI models are viewed as the top performers.
Polymarket currently ranks Google’s Gemini at the top, followed by OpenAI’s ChatGPT and Elon Musk’s Grok. But these standings are based on Arena benchmark scores — not revenue or usability.
For business le
Niv Nissenson
Jul 223 min read
bottom of page