Insights on Belgian tax law and AI-powered tax research from Auryth.
The era of making models bigger is ending. Five paradigms are replacing brute-force scaling — hybrid architectures, inference-time reasoning, world models, self-improvement loops, and hardware co-design. What each means for AI in regulated industries.
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.
$8 billion. $1.8 billion. $133 million. The world's best-funded legal AI companies are proving the model works. They're also proving what it can't do for Belgian tax.
79% of legal professionals now use AI. Tax firms tripled adoption in one year. But the real story isn't the average — it's the widening gap between strategic adopters and everyone else.
Harvey raised $760M. Blue J raised $133M. Neither can answer a question about erfbelasting in Brussels. Here's why that creates an opportunity.
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.
Accuracy claims without published metrics are marketing. Here's what it takes to measure legal AI honestly, and why the industry avoids it.
When AI-assisted tax research leads to wrong advice, Belgian law is clear: the professional pays. But that's actually the strongest argument for using transparent AI — not against it.
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.
Stanford found that premium legal AI tools hallucinate 17-33% of the time. But the most dangerous finding wasn't the hallucination rate — it was misgrounding.
First-stage retrieval finds 100 matches. Reranking identifies the 5 that actually answer your question. Here's why that distinction matters for legal AI.
Keyword search finds exact article numbers. Semantic search finds related concepts. Hybrid search does both — and the difference is measurable.
Most AI implementations fail because firms skip the trust-building phase. A practical 4-phase framework for Belgian tax practices.
A couple in Antwerp with €2M in assets. Two children in different regions. Three inheritance tax regimes. Here's how the numbers diverge — and what a complete advisory analysis actually looks like.
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.
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 actually means for Belgian tax professionals — and what it doesn't.
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.
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.
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.
Belgium has three inheritance tax regimes, three gift tax structures, and three registration duty rates — each with different reform timelines. Here's what that means for advisors and why no single answer is ever complete.
Belgian tax law changes twice a year minimum. If your AI tool can't tell 2019 from 2026, its answer isn't wrong — it's for the wrong year. Here's what temporal versioning is and why it matters.
A single TAK 23 life insurance product touches insurance tax, income tax, withholding tax, TOB, and regional inheritance tax. Here's what most advisors — and every AI chatbot — miss.
Chat interfaces feel modern but produce ephemeral, indefensible output. Professional tax research needs structured answers you can file, reproduce, and defend.
A ministerial circular and a Court of Cassation ruling look the same to most AI search systems. That's not a minor flaw — it's a fundamental gap that makes every answer unreliable.
AI skepticism in tax is rational. Most tools deserve it. But rejecting the entire category because ChatGPT hallucinated a tax rate is like refusing calculators because the first ones jammed.
These aren't trick questions. They're routine queries any tax professional faces. But they expose five architectural blind spots that no general-purpose AI can fix with a better prompt.
How retrieval-augmented generation works, why basic RAG still hallucinates, and what a search-RAG fusion architecture adds for tax professionals.
The AI industry obsesses over accuracy benchmarks. For tax professionals, verifiability is the metric that actually protects you.
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 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.
LLMs overestimate their own correctness by 20-60%. Confidence scoring doesn't fix that problem — it makes it visible. For tax professionals, that visibility is the difference between a research tool and a guessing machine.
Three tax questions, two AI tools, one clear lesson: for professional research, verifiability beats confidence.
Why language models invent legal citations, what makes Belgian tax especially vulnerable, and three defenses that actually work.