peony
Equity Investor & Researcher
Wiley's sell-off is a gift: it is one of the most promising AI data players, trading under 2x sales while projecting 27% growth, yet the market dumped it on one flat quarter. Peer-reviewed journals carry some of the highest per-token value of any data, which should rise dramatically in a few years, so taking on leverage to acquire content is welcome. @Heady_Creek counters that the core business is declining and AI growth comes from acquisitions like Emerald. x.com x.comÂ
John Wiley & Sons is only in inning two as an AI data play, with data valuable to AI, aggressive AI adoption and a conservative valuation. Top data partners should earn orders of magnitude more once AI data centers generate real value, and more AI means more scientific publications, a feedback loop. Peer review guards against AI consuming its own slop, and real-time data keeps contracts steady or growing. Its role as a clinical data broker to AI could grow to be worth today's entire market cap. x.com x.comÂ
Accenture and Wolters Kluwer benefit massively from cheaper AI tokens and margin gains as they deploy AI. Their knowledge costs rank among the highest relative to COGS, but regulatory capture, reputation and sticky customers keep revenue broadly intact. Neither shows clear signs of slowing growth; Accenture guided to high-single-digit revenue growth from 2029. Despite enormous terminal-value compression, an unwinding AI-infrastructure trade could bring a sizeable rerating if fundamentals hold. … Show more
AI models are becoming a commodity, so the moat is moving to the data layer, which the market has not repriced. Innodata creates training data; Reddit licenses human conversation to Google and OpenAI at a 0.76 PEG; Snowflake stores it with 126% net retention; ServiceNow deploys agents, with $1M+ Now Assist customers up 130%; CrowdStrike secures it at 97% gross retention; Datadog bills more as inference scales. Palantir grows fastest but at ~67x forward sales. Reddit and Datadog are best value. … Show more
Reddit is well placed as the source every AI model wants to license training data from, while its ad revenue booms. Q2 revenue rose 61% to $804.9 million, operating income more than tripled to $231.7 million and net income reached $252.8 million. At about 17 times two-year forward earnings, the stock isn't expensive. x.com x.comÂ
Reddit's selloff over a possible end to its Google AI data deal is overdone, because Reddit owns content AI models depend on and has leverage to win usage-based pricing. Per WSJ, the roughly $60M-a-year contract has not been cancelled; it is 1.4% of 2026 consensus revenue, yet $3B of value was erased. Reddit ranks at or near the top of AI citation studies, and a UK CMA ruling lets it withhold training data while keeping search referrals. The July 30 earnings should clarify the renewal. x.com … Show more
Wolters Kluwer may be a massive opportunity, with a huge margin of safety assuming 8% annual free-cash-flow growth and a terminal P/E of about 15. x.comÂ
Snowflake is overlooked AI infrastructure because every enterprise AI model needs clean, governed data to run on. It has 126% net revenue retention, $9.77 billion of contracted future revenue and 813 of the world's 2,000 largest companies as customers. x.comÂ
