What does POET actually do?

POET Technologies Inc. makes tiny chips that use light instead of electricity to move data around. Think of it like replacing copper water pipes with fiber optic "light pipes" inside the massive data centers that power AI (like ChatGPT, image generators, and cloud computing). Their key invention, the Optical Interposer, is like a special adapter that connects electronic parts and light-based parts on a single tiny chip. The company is based in San Jose, California, listed on NASDAQ, and is still in the early stages of making money — they have $5.6M+ in confirmed orders and over $300M in the bank.

The Big Picture: From Almost Nothing to $91M?

POET went from earning just $3,700 in Q3 2024 to $298,000 in Q3 2025 — that is an 80x increase. Still tiny, but it shows the direction is up. They have $5.6M+ in confirmed orders to deliver in 2026. Analysts estimate they could reach $91M in revenue in 2026, which would be a massive jump. The $300M+ cash they have means they can afford to wait and build without worrying about running out of money. The key question: can they actually deliver all those orders on time and at scale?

Reading the Losses: It Is Not All Bad News

Yes, POET keeps losing money every quarter — but look at the direction. The Q3 loss shrank to -$9.4M from -$17.3M in Q2. Their cash spending also went from -$7.7M (Q2) down to just -$2.8M (Q3). The losses are getting smaller while revenue is getting bigger. Most of the spending goes to research and development — meaning they are investing in building new products. The company expects to grow revenue steadily throughout 2026 as they ship the $5.6M+ worth of orders. The big question is: will revenue grow fast enough to cover the losses?

How it works (Think of it like baking)

The traditional way of making optical components is like decorating cupcakes one by one by hand — slow, expensive, and hard to scale. POET's Optical Interposer is like using a cookie cutter on a big sheet of dough: they make thousands of chips at once on a large silicon wafer, then cut them apart. This "wafer-level" approach gives POET three big advantages:

• 50%+ cheaper — Automated factory production instead of hand-assembly
• 5x smaller — Everything fits on one tiny chip instead of a bulky module
• Uses less power — Fewer connections means less wasted energy and heat
• Grows easily — The same design works at 400G, 800G, 1.6T, and 3.2T speeds

The Good News: POET Will Not Run Out of Money

With $300M+ in cash and spending about $10-12M per quarter, POET has enough money to operate for over 5 years even if they never earn another dollar. This is extremely important for a young company. Many startups fail not because their technology is bad, but because they run out of cash before they can sell enough. POET does not have this problem. The $150M fundraise in October was oversubscribed (more investors wanted in than there was room for), which shows big institutional investors believe in the story. The question is no longer "Can POET survive?" but "Can POET turn all this cash into a real, growing business?"

Why Partners Matter So Much

When you see names like Mitsubishi Electric (a $35B+ Japanese electronics giant), Semtech ($5B+ company), and Foxconn (the world's largest electronics manufacturer — they build iPhones) working with a small company like POET, it tells you something important: these massive corporations did their homework and decided POET's technology is real and worth investing time in. Big companies do not partner with companies they think are fake. This does not guarantee POET will succeed, but it greatly reduces the chance that the technology is a scam.

What Is Happening Right Now on the Chart?

POET is in a "cooling off" period after its big fundraise. When a company sells new shares at $7.25, people who bought them are currently underwater (losing money) at $5.58. This creates a "ceiling" because those buyers will likely sell to break even when the price gets back to $7.00-$7.25. On the bright side, the 200-day average at $4.80 acts as a long-term floor. For the price to break above $7.00, POET needs a positive catalyst — like a major new order, a great Q4 earnings report, or a new partnership announcement.

Why should you care about this?

This is the single most important thing to understand about POET. The entire reason this company could be worth much more in the future — or remain a niche player — comes down to one massive trend: the world's biggest tech companies are being forced to replace copper wires with light inside their data centers. If this switch happens as fast as expected, companies like POET that make optical components will be in very high demand.

Think of it like this...

Imagine two delivery systems for a warehouse. Copper is like using golf carts: they work great for short trips inside a single building, but they are slow, can only carry small loads, and overheat if you push them too hard. Light (optical) is like using a high-speed rail system: it can carry enormous loads over long distances with barely any energy. As warehouses (data centers) get bigger and need to move more stuff (data), golf carts (copper) simply cannot keep up. That is why everyone is switching to high-speed rail (optical).

POET Makes Engines, Not Cables

One important detail: POET does not make cables or complete transceiver modules. They make the "engine" — the most valuable, high-tech part inside those modules. Other companies (LuxshareTech, Adtran) then build the complete product around POET's engine. Think of it like this: POET designs and builds the engine for a race car, but does not build the car itself. This is a high-margin, IP-focused business model similar to ARM (which designs processor blueprints used in most smartphones, but does not make the phones). The market POET is targeting — optical engines for AI — is estimated to be worth $3 to $5 billion by 2028.

David vs Goliath

POET ($827M) is a tiny fish in a giant ocean. NVIDIA alone is 3,800 times larger. Even Coherent is 18 times bigger. But here is the opportunity: if POET captures just 1% of the optical engines market ($3-5 billion by 2028), that would mean $30-50 million in revenue — a massive increase from today's $1.2M. POET's advantage is that it does not try to compete head-on with the giants. Instead, it provides a unique technology (the Optical Interposer) that the giants' customers need. The risk? A company like Broadcom or Intel could develop something similar in-house and shut POET out.

Most Likely Good Buy Zone: $4.80 to $5.21

Based on technical analysis, the $4.80 to $5.21 range is where the stock is most likely to find its bottom (about 60% probability). This zone has multiple layers of "floor" support: the 200-day average, the uptrend line, and an area where a lot of shares have changed hands before. If you are considering buying, this range offers the best risk-to-reward ratio. The danger zone is below $4.20 — if the stock falls that far, it could mean the business story is failing and you should reconsider. A gradual approach (buying some now, some lower) can help manage risk.

The Idea Behind This Trade

POET is a high-risk, high-reward bet on light-based technology for AI. The technology has been validated (industry award, major partners like Mitsubishi and Semtech), and the company has $300M+ in cash, so it will not disappear. The best time to buy is in the $5.00 to $5.50 zone where the stock is consolidating. Target 1: $7.30 (the average analyst price target, about +33% gain) could be reached on good Q4 earnings or a new big order. Target 2: $8.50 (+54%) is achievable if 2026 revenue is on track. The stop loss at $4.20 limits your maximum loss to about 24%. Key dates to watch: Q4 earnings (March 30), first shipments in Q1 2026, and any announcements about 1.6T Teralight orders.