Meta Platforms employees have begun testing a prediction-market app that is still under development, The Wall Street Journal reported on Friday, citing people familiar with the matter. The paper called internal testing “a step the company often takes ahead of a public release”. On the app, currently called Takes, employees wager on the outcome of questions or events using “cred”, short for credibility, the people said.
Unlike Kalshi and Polymarket, Meta’s app does not let users wager real money, the Journal reported. Kalshi, the larger of those two, said on 29 October that it had raised $1 billion at a $40 billion valuation and that over 30 million Kalshi accounts had been made. Meta reported 3.64 billion daily active people across its apps in September, about 121 times Kalshi’s account count, by our arithmetic; the two are different measures.
Business Pill · Distribution Advantage Explained
An evergreen Business Engineer explainer on distribution advantage, the general idea behind a company that can put a new product in front of an existing audience. It teaches the concept only and does not discuss Meta or Takes.
The key insight: As we read the WSJ report next to Kalshi’s own figures, Meta is testing the prediction-market format without the part that makes Kalshi a regulated money business: wagers in cred, not cash. What Meta brings is reach; its September audience is about 121 times Kalshi’s account count, by our arithmetic, though the measures differ.
The WSJ article, by Meghan Bobrowsky, was published at 2:49 p.m. Eastern on 13 November, 19:49 UTC, and marked as an exclusive. We could read its first two paragraphs; the rest is behind the paper’s paywall. Those paragraphs give the name, Takes, the currency, cred, and the test group, Meta employees.
The Journal said employees can wager on “the outcome of questions or events”. It contrasted the app with “other prediction markets operated by Kalshi and Polymarket”, saying Meta’s app “doesn’t allow users to wager real money”. The portion we read gives no launch date and no Meta comment.
Each square stands for 30 million. Meta reported 3.64 billion family daily active people on average for September 2026 in its third-quarter results; Kalshi said on 29 October 2026 that over 30 million Kalshi accounts had been made. The two are different measures; the ratio of about 121 is our arithmetic.
Where Takes Sits Next to Kalshi
Kalshi’s funding announcement of 29 October described “a $1 billion Series G at a $40 billion post-money valuation, co-led by Wellington and Sequoia”. It said: “Over 30 million Kalshi accounts have been made, and we account for over 75% of prediction market trading volume”.
Kalshi also said its “overall business’s annualized trading volume now exceeds $1 trillion”, with crypto volume at $140 billion annualized and politics above $8 billion. These are the company’s own figures. We have covered Kalshi prices before, as in our piece on Anthropic IPO odds on Kalshi.
Meta’s reach.Meta’s third-quarter results put family daily active people at 3.64 billion on average for September 2026, up 3% year over year. Divided by Kalshi’s 30 million accounts, that is about 121 times, by our arithmetic. Kalshi counts accounts ever created; Meta counts people using its apps each day, so the ratio shows scale, not a like-for-like user comparison.
Kalshi’s annualized trading volume by line, as stated in its funding announcement of 29 October 2026. These are the company’s own figures.
Meta’s Earlier Plans
In June, CoinDesk, summarising a New York Times report, said Meta was developing an experimental app called Arena that would let users forecast outcomes in politics, sports, entertainment and world affairs. CoinDesk said Arena was expected to use “a video game-like points system rather than cash”, though Meta “has not ruled out eventually allowing real-money betting”.
The part of the WSJ article we read does not say whether Takes is the app the Times called Arena. Meta’s other consumer AI work this week includes its Muse assistant; our piece on Muse in Chrome has the figures.
The Structural Read
As we read it, a points currency keeps Takes closer to a social feature than to an exchange: the WSJ says users wager on questions or events, and Kalshi’s announcement describes its business as an exchange with fees, institutions and futures.
As we read the June and November reports together, the Times described Arena as points-based with real money not ruled out, and the WSJ describes Takes as cashless in employee testing; neither report, as we read them, gives a public release date. Our piece on Polymarket’s Anthropic IPO odds shows how these markets price AI news.
Kalshi, funding announcement, 29 October 2026
“Over 30 million Kalshi accounts have been made, and we account for over 75% of prediction market trading volume.”
Three Implications
FOR KALSHI AND POLYMARKET The WSJ names both as the real-money markets Takes differs from. Kalshi says it was valued at $40 billion on 29 October, with over $1 trillion of annualized trading volume.
FOR META WATCHERSMeta reported 3.64 billion daily active people for September. Our piece on Muse in Chrome covers Meta’s other consumer AI product this week.
FOR READERS OF THE COVERAGE The WSJ details come from people familiar with the matter; the portion we could read carries no Meta comment.
The framework defines its distribution layer as “The surfaces through which AI reaches end users.”
As we read it, Takes is Meta testing one more surface on top of an audience of 3.64 billion daily people, where Kalshi has had to build its own. The framework does not discuss prediction markets or Takes.
What Is Not Established
We could not read the full WSJ article, so we do not know whether it carries a Meta comment, a planned release date, which questions employees can wager on, or how cred is earned. We do not know whether Takes and Arena are the same project.
Kalshi’s figures are its own and are not audited in what we read. Meta’s daily active people and Kalshi’s accounts measure different things. We did not contact Meta, Kalshi or the WSJ for this piece.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Meta employees are testing Takes, a prediction-market app that uses a points currency called cred rather than money, the WSJ reported, citing people familiar with the matter. Kalshi, which says it handles over 75% of prediction-market trading volume, was valued at $40 billion last month with over 30 million accounts. Meta’s September audience of 3.64 billion daily people is about 121 times that account count, by our arithmetic.
A note on sourcing. The Wall Street Journal article was behind the paper’s paywall; we read its headline, standfirst and first two paragraphs, which carry every WSJ detail in this piece. Kalshi’s figures come from its own funding announcement of 29 October 2026 and Meta’s from its third-quarter results release, both read in full. The Arena report is the New York Times’s, as summarised by CoinDesk on 23 June 2026.
The ratio of Meta’s daily active people to Kalshi’s accounts is our arithmetic. We did not contact Meta, Kalshi or the WSJ for this piece.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
