Learn · a trick with a name
Headline Over Filing
the document underneath says something else.
Press releases are marketing. Filings are legal documents. When they disagree, the filing wins.
How to spot it
- A company's announcement and its filing use different numbers
- Risk factors in the filing that the press release skips
- Journalists quote the release, not the document
The one question to ask“What does the filing say?”
At work
For any funding, IPO or earnings claim that matters to you, read the filing's first pages. The caveats live there.
Caught in the wild
Every time we've caught it so far. Try calling each one before you open it.
- TRUE, BUTThe $16 billion in Meta scam-ad money is real, but it is a year-old projection that counts banned goods too. The fraud-only figure in the same documents is $7 billion.
- TRUE, BUTMicrosoft's patent really describes AI that swaps a speaker's words and redraws their lips. But its first claim edits one viewer's copy, from that viewer's profile, and the application has drawn examiner objections.
- TRUE, BUTNikkei says Big Tech's hidden AI debt nearly doubled to $3.2 trillion. The relays we could read describe leases and purchase contracts in footnotes, and another tally doubles Meta's share.
- TRUE, BUTA headline says Postman tries to share all your API data. The GitHub report it summarises documents a silent plugin install and says it does not establish data exfiltration.
- TRUE, BUTTom's Hardware's headline says sweeping new Claude restrictions block game cloning and reverse-engineering. Anthropic confirms a game rule, and its own 6 October post says its public models block most cyber work.
- TRUE, BUTThe Dutch are cutting 25,000 civil service jobs in an AI-led overhaul, says the headline. The government calls 25,000 an estimate tied to savings NL Times puts at 3.5 billion euros. Its AI targets arrive in 2027.
- TRUE, BUTApricitas counts more than 200k lost creative jobs and asks how much is AI. Its own answer: proving the exact amount is extremely difficult.
- TRUE, BUTScientists spotted AI, a contest winner lost his title. Our record names one watermark check, and Nikon's statement does not say which element broke the rule.
- TRUE, BUTOne hacker, 100+ companies, $50 each. The $50 is an estimate built from a $5-per-task guess and a 10% success rate, and no bill appears in the record.
- TRUE, BUTThe headline says OpenAI's agents switched off their own watchdog. OpenAI's report says agents weakened monitoring in two coding jobs, and in one the monitor kept running.
- TRUE, BUTWhatsApp's help page on deleting your account told its support chatbot: do not recommend contacting any Data Protection Authority. WhatsApp's own privacy policy names one.
- TRUE, BUTA headline says Anthropic's safety bill only kicks in at 50 deaths or $1 billion in damage. Those figures are a risk test for blocking a release, and California law has used the same line since 2025.
