#legal AI
21 posts
Bilingual Belgian tax law: why NL/FR is a prerequisite, not a feature
Federal law is published in Dutch and French. Regional codes exist in one language only. Rulings follow the language of the procedure. A monolingual AI tool misses half the corpus.
How we handle contradictory sources, and why most AI tools don't
When Belgian tax sources disagree, the worst thing an AI tool can do is pick one and act confident. Here's what honest uncertainty looks like.
Why we publish our accuracy, and why almost nobody else does
Accuracy claims without published metrics are marketing. Here's what it takes to measure legal AI honestly, and why the industry avoids it.
Behind the scenes: how a law change flows through a legal AI system
When Belgium's programme law amends 47 provisions overnight, what happens inside the AI tool you rely on? A transparent look at the ingestion pipeline.
What the Stanford hallucination study actually revealed, and why the industry's response missed the point
Stanford found that premium legal AI tools hallucinate 17-33% of the time. The most dangerous finding was misgrounding.
What is reranking, and why it's the difference between finding documents and finding answers
First-stage retrieval finds 100 matches. Reranking identifies the 5 that actually answer your question. Here's why that distinction matters for legal AI.
How hybrid search works, and why your legal AI tool probably uses only half the equation
Keyword search finds exact article numbers. Semantic search finds related concepts. Hybrid search does both, and the difference is measurable.
The EU AI Act and legal AI: what Belgian tax professionals actually need to know
The AI Act entered into force on 1 August 2024. Some provisions are already active. Most articles confuse provider obligations with deployer obligations. Here's the distinction that matters for your practice.
The future of tax research: what AI changes, what it doesn't, and what that means for your practice
AI adoption in accounting firms surged from 9% to 41% in one year. McKinsey says 44% of legal tasks are technically automatable. The World Economic Forum lists bookkeeping among the fastest-declining roles. Here's what this means for Belgian tax professionals, and what it doesn't.
What is a knowledge graph, and why it changes how AI navigates Belgian tax law
A search engine finds text. A knowledge graph navigates relationships: which article amends which, which ruling interprets what, which exception overrides the rule. Belgian tax law is a web of cross-references, and a knowledge graph is the map.
What is chunking, and why it's the invisible foundation of legal AI quality
Before your legal AI tool can answer a question, it has to cut the law into pieces. The way it cuts determines whether the answer includes the exception that changes everything, or misses it entirely.
How much time does tax AI actually save? An honest estimate
Harvey users report 37 hours per month. Thomson Reuters claims 63% reduction. We looked at the real numbers for Belgian tax research, and built a conservative estimate that doesn't require you to believe marketing claims.
What is temporal versioning, and why your legal AI tool probably serves you yesterday's law
Belgian tax law changes twice a year minimum. If your AI tool can't tell 2019 from 2026, its answer may be correct for the wrong year. Here's what temporal versioning is and why it matters.
Why we're not building a chatbot
Chat interfaces feel modern but produce ephemeral, indefensible output. Professional tax research needs structured answers you can file, reproduce, and defend.
What is RAG, and why it's not enough for legal AI on its own
How retrieval-augmented generation works, why basic RAG still hallucinates, and what a search-RAG fusion architecture adds for tax professionals.
Why transparency matters more than accuracy in legal AI
The AI industry obsesses over accuracy benchmarks. For tax professionals, verifiability is the metric that actually protects you.
How to evaluate a legal AI tool: 10 questions that actually matter
Most firms start by asking 'how accurate is it?' That's question 10 on this list. Here are the nine questions you should ask first, and why they matter more for professional tax work.
Fine-tuning vs. RAG: two ways to make AI smart, and why it matters which one your tax tool chose
Fine-tuning memorizes yesterday's law. RAG looks up today's. For Belgian tax professionals, this architecture choice determines whether your AI tool is current or confidently outdated.
What is confidence scoring, and why it's more honest than a confident answer
LLMs overestimate their own correctness by 20-60%. Confidence scoring does not fix that problem. It makes the problem visible, and for tax professionals that visibility is the difference between a research tool and a guessing machine.
I asked ChatGPT and Auryth the same Belgian tax questions. Here's what happened
Three tax questions, two AI tools, one clear lesson: for professional research, verifiability beats confidence.
AI hallucinations: why ChatGPT fabricates sources (and how to spot it)
Why language models invent legal citations, what makes Belgian tax especially vulnerable, and three defenses that actually work.