Oliver | MMMT Wealth
Founder of MMMT Wealth, CPA, Technical/Fundamental Analysis
POET Technologies has little upside left for the risk after rising 85% in three trading days. It has a real $5.6M production order, a 1.6T optical receiver with Semtech ramping this year and $313M in cash. But even at 12x forward sales, like Applied Optoelectronics, on $111M of expected FY27 revenue, its value roughly equals today's $1.6B market cap, and 15x gives only about $2B. Upside makes sense only at 30x or more, which is possible, as Sivers shows. x.com x.com x.comÂ
Applied Optoelectronics' possible 2028 revenue makes even a 100% takeover premium too small if it delivers on 2027-28 expectations. Revenue of $6-8B in calendar 2028 at an illustrative 8x multiple would imply about $56B in market value, versus $7B today. That is the upside worth holding out for. x.comÂ
These high-flying stocks look like they are running purely on hope, with valuations hard to justify. POET Technologies trades at 1,580x sales, D-Wave Quantum at 874x, Navitas Semiconductor at 169x and AXT at 96x, while Oklo carries an $11.4B market cap with no sales. x.comÂ
POET Technologies' 47% one-day plunge was self-inflicted, not a technology problem, making it an easy buy. Marvell, which bought Celestial AI in February, pulled Celestial's orders citing a confidentiality breach after POET's CFO confirmed the order on Stocktwits. The stock has since rebounded 211.67% in two weeks. x.comÂ
Credo's near-total share of AI data-center cables and unchanged estimates make its lower multiple look like a fear-driven opportunity. Its stock went from 27x to 20x forward earnings while the supply chain building those systems guided 2027 revenue to double. Buyers this week will probably be very happy a year from now. x.comÂ
POET Technologies' leadership is very good at hyping the stock, with two such cases in three weeks. It disclosed a new $5 million contract in the SEC filing that announced Marvell's order cancellation, and headlines ran with a "$500 million" Lumilens order that its own release put at $50 million with potential to scale. It is also oddly letting Lumilens buy deep in-the-money POET warrants at greatly reduced prices, not tied to a 30-day VWAP formula, unlike AMD's OpenAI deal. x.com x.com x.comÂ
POET Technologies is the upstream pick-and-shovel for Celestial AI's Photonic Fabric, which Marvell just acquired for about $5B. POET supplies the light source that powers it, and Marvell has the hyperscaler relationships with AWS, Google and Meta to deploy it at scale. Revenue is still early, but once Marvell integrates Celestial and ramps production orders, the market should recognize POET's value; Q3 or Q4 2026 could be the inflection, and patience is required. x.comÂ
POET Technologies' market cap rose $850M on a $50M deal, a jump that would require 1,700% margins on the product to justify. x.comÂ
POET Technologies looks wildly overvalued at 1,308 times trailing sales. At the March 2000 dot-com peak, the richest notable surviving stocks traded far lower, with Cisco at 35x, Oracle at 27x and Microsoft at 26x. x.comÂ
POET Technologies has never really had a functioning product, and its lack of revenue speaks for itself. Unlike Bloom Energy, which had a working product before data-center demand arrived, POET has little to show after all these years. Management keeps cashing out instead of keeping real skin in the game. It could still have its day, but treating it as a sure thing was wrong. x.com x.com x.com x.comÂ
POET Technologies is a weak company because Celestial AI and Marvell are not going to use its technology. x.comÂ
