Members read every story in the archive, re-verified. Sources and corrections stay open to everyone.
What the pen marks mean
rememberThe takeaway: what to remember
figureThe number that matters
evidenceThe evidence line
rulingThe ruling
claimThe claim that fails
Each mark is drawn once, as you reach it. Numbers are only circled when they come from the story’s own figures.
01 / RulesBS
Medicare's number two said AI prior-auth contractors don't earn more by denying care. Medicare's own page says they get a cut of care averted.
Chris Klomp, CMS deputy administrator and nominee for HHS deputy secretary ·
WISeR uses AI and machine learning, with human clinical review, to decide prior authorization requests for selected procedures in six states. At a Senate HELP hearing, Senator Patty Murray asked Klomp whether its contractors make more money if they deny care. Ars Technica reports he replied that his understanding was no.
Members · 30 days free
See what the evidence actually shows.
Members read the full check on every story: what the evidence shows, why it matters to you and the one line to take with you. Every past edition, re-verified, and the Receipts Pack come with it. A$89 a year, about A$0.24 a day.
First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge.
Open the evidence3 source pages +
The claim we checked
Asked at a Senate hearing whether contractors in WISeR, Medicare's AI-assisted prior authorization pilot, make more money if they deny care, CMS deputy administrator Chris Klomp answered 'My understanding is no' and said inappropriate denials carry significant financial penalties.
These are quoted receipts, not a count of independent investigations. Several reports may rely on the same original source.
Do the contractors in the model—who are the private companies conducting the prior authorization assessments—make more money if they deny care? Just yes or no?
CMS documents written as a guide for WISeR participants explain further that for every denied request, CMS will determine what the regional benchmark cost for that care would have been and then pay the company 25 percent.
WISeR will run for six performance years from January 1, 2026 to December 31, 2031 in six states: New Jersey, Ohio, Oklahoma, Texas, Arizona, and Washington.
The Pentagon says an appeals court 'completely' validated its Anthropic blacklisting. It won on one of two designations.
Sean Parnell and Pete Hegseth, Department of War ·
The D.C. Circuit denied Anthropic's petitions against its exclusion under a supply chain security law in a 2-1 decision. The majority said the Department had ample support to treat Claude's built-in restrictions as a national-security risk and rejected Anthropic's constitutional claims. Judge Henderson dissented.
Members · 30 days free
See what the evidence actually shows.
Members read the full check on every story: what the evidence shows, why it matters to you and the one line to take with you. Every past edition, re-verified, and the Receipts Pack come with it. A$89 a year, about A$0.24 a day.
First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge.
Open the evidence4 source pages +
The claim we checked
After the D.C. Circuit ruled against Anthropic, Defense Department spokesman Sean Parnell said the ruling 'completely validates the Department's position', and Pete Hegseth posted 'Confirmed: @AnthropicAI = Supply Chain Risk'.
These are quoted receipts, not a count of independent investigations. Several reports may rely on the same original source.
We reject these challenges. The Department had ample support for its conclusion that the continued integration of Claude into the Department's information systems, by the Department or its contractors, presented a statutorily covered national-security risk.
Another federal court has already held the government's parallel designation unlawful. We remain confident in our position and are considering all options, including further review,
'DeepSeek $1B ARR' is a run rate from unnamed sources. The same reporting puts seven months of revenue at about $70.7 million.
Liang Wenfeng, DeepSeek CEO (per The Information), amplified by TLDR AI ·
The Information reported, citing unnamed sources, that DeepSeek's annualized revenue run rate reached $1 billion, and PYMNTS relayed that the CEO shared the figure with investors. It followed price increases of 2.3 to 4.5 times. DeepSeek did not reply to PYMNTS.
Members · 30 days free
See what the evidence actually shows.
Members read the full check on every story: what the evidence shows, why it matters to you and the one line to take with you. Every past edition, re-verified, and the Receipts Pack come with it. A$89 a year, about A$0.24 a day.
First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge.
Open the evidence3 source pages +
The claim we checked
DeepSeek's annualized revenue run rate has more than doubled to $1 billion, a figure CEO Liang Wenfeng reportedly shared with investors; TLDR AI headlined it 'DeepSeek $1B ARR'.
These are quoted receipts, not a count of independent investigations. Several reports may rely on the same original source.
DeepSeek more than doubled its annualized revenue run rate over the past few months, bringing the rate to $1 billion, The Information reported Wednesday (Sept. 23), citing unnamed sources.
The company generated roughly 475 million yuan (about $70.7 million) in the first seven months of the year, or about 10 times its revenue for all of last year, the report said, citing unnamed sources.
The report said that DeepSeek raised the prices of its models by 2.3 to 4.5 times, but that the company’s prices remain among the lowest for major AI models.
Latest update reveals China's AI firm DeepSeek has doubled its annualized revenue run rate to reach $1 billion, up from under $500 million just months ago.
Claude agents matched people's book tastes on 61% of pairs. Anthropic says a coin flip gets 50%.
Anthropic (Project Swap), amplified by TLDR AI ·
In Project Swap, 201 Anthropic employees had a short chat with Claude, then sent Claude agents to trade books for them. Each person also ranked 10 books, which the agents never saw, so Anthropic could score how well each agent understood its person.
Members · 30 days free
See what the evidence actually shows.
Members read the full check on every story: what the evidence shows, why it matters to you and the one line to take with you. Every past edition, re-verified, and the Receipts Pack come with it. A$89 a year, about A$0.24 a day.
First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge.
Open the evidence2 source pages +
The claim we checked
Anthropic says that from a five-minute chat, a Claude agent's ranking of books matched its person's on 61% of pairs, 'surprisingly good for such a short conversation', in a book-trading market for 201 Anthropic employees.
These are quoted receipts, not a count of independent investigations. Several reports may rely on the same original source.
From a five-minute chat, an agent’s ranking of the books matched its person's on 61% of pairs, which is surprisingly good for such a short conversation.
Taking this into account, the best possible assignment (the utilitarian optimum) in our experiment is a score of 0.89 overall, leaving participants at roughly their second choice on a 10-book list.
So, working from Claude’s imprecise rankings accounts for a majority (85%) of the shortfall, and sending agents into a “free-for-all” trading floor accounts for the remaining 15%.
So this summer, we built a small, controlled market to study these questions: a barter economy with 201 Anthropic employees and their Claude-powered agents.