Chips&Media (094360) Stock Outlook 2026: Video Codec IP Royalties and the Small-Cap Dilemma

The Core Question Before Buying Chips&Media

The whole case for Chips&Media compresses into a single tension: is this a high-margin, asset-light IP company that happens to be small, or a small-cap component supplier whose earnings ride on other people’s chip shipments? Both descriptions are true. Where you place the weight decides both the multiple you’ll pay and how you’ll trade it.

My view up front. On business model alone, Chips&Media belongs to the same lineage as ARM and Ceva — a genuinely attractive royalty company. It deploys almost no capital, and a block of IP licensed once can ride into multiple chips and tens of millions of devices, stacking royalties at near-zero incremental cost. That is a good business. But you have to hold that alongside the reality: it is small, concentrated in the narrow video-codec domain, and exposed to the consumer-electronics, automotive, and security-camera cycles at its front end. Good model, small scale — investors who see only one of those two faces tend to get hurt.

Buy this as a “semiconductor IP, therefore automatic hypergrowth” theme and you’ll be shocked by the drawdown when the royalty cycle rolls over. Dismiss it as “just another tiny KOSDAQ part” and you’ll miss the earnings explosion when royalty leverage actually kicks in. That classification is where returns are won or lost.

👉 For a comparison with a materials play in the same semiconductor value chain, read our ENF Technology (102710) stock outlook 2026.

---

The IP Licensing Model: Collecting Royalties Without a Fab

The starting point for understanding Chips&Media is that it does not make chips. It doesn’t etch wafers or assemble packages. What it sells is a design asset — a hardware implementation of video codecs delivered as blueprints and code.

Revenue runs on two rails.

License fees. When a customer — an SoC or chip design house — commits to putting the company’s video encoder or decoder IP into a chip, an upfront license fee books at that moment. This is the seed. It’s cash today, but its real value lies in what it harvests years later.

Royalties. Every time a chip containing the licensed IP is manufactured and ships into the market, Chips&Media collects a set amount per chip. A phone, TV, car, or camera sold means a royalty from the chip inside it. This is the fruit. The gap between signing a license and actually collecting royalties usually runs several years, because the pipeline — chip design, verification, mass production, device integration — is long.

That time lag is what makes IP-company investing hard. Strong royalties this year mean licenses seeded years ago are now bearing fruit; a jump in new licenses today is a reservation for royalties years out. So the investor always has to watch two clocks at once.

Keep the contrast with a foundry sharp. A foundry takes someone else’s design and prints physical wafers — a capital-heavy manufacturer where depreciation and fab utilization drive the numbers. IP licensing is the mirror image: no fab, so little capital, and once royalties begin there’s almost no incremental cost. Beyond headcount and R&D there isn’t much variable cost, so past a certain scale profit compounds much faster than revenue. That operating leverage is the real draw of the IP model.

---

Royalty Leverage: The Appeal and the Double-Edged Sword

Let’s make the leverage concrete. Say a codec IP developed once goes into three of Customer A’s chips, and those chips each ship in tens of millions of devices. Chips&Media’s incremental cost is effectively nil. The IP already exists; royalties accrue automatically in proportion to units shipped.

Stage	What happens	Company benefit
New license signed	Customer adopts the IP	Upfront license revenue (the seed)
Customer chip ramps	Chip ships into the market	Royalties begin (first harvest)
Spread across chips and devices	Cumulative shipments rise	Royalties stack at zero added cost
Resell new IP to same customer	Codec plus NPU plus ISP bundle	Higher royalty value per customer

The beauty of that table is that the right column can grow far faster than the left. But the reverse holds just as firmly. When front-end chip shipments roll over, you cannot defend royalties by cutting costs — if the product doesn’t sell, the royalty doesn’t arrive. Fixed-cost R&D and payroll stay put while royalties shrink, so profit falls faster than revenue. Leverage runs both ways.

Now layer on the small-cap condition. Whether one or two customers’ specific chips sell well can swing an entire quarter. Unlike a large IP firm diversified across hundreds of customers and thousands of chips, Chips&Media carries high revenue concentration and is sensitive to the fate of individual projects. Royalty leverage is explosive in the good phases and correspondingly painful in the bad ones.

👉 To frame the broader semiconductor and AI value chain at the stock level, see our AI Stocks Investment Guide 2026.

---

Is the Moat Real? The Defensibility of Codec IP

Granting that the licensing model is attractive, how durable Chips&Media’s moat actually is remains a separate question. Break it into honest layers.

Standards support and verification assets. Video codecs follow international standards — H.264, HEVC, VP9, AV1. The standards themselves are public, but implementing them as low-power, high-efficiency hardware and verifying compatibility and quality across a mass of real devices is another matter. Chips&Media has a long track record of shipping multiple codec IP generations in commercial silicon. For a customer, buying verified IP reduces the risk of a chip respin.

Switching costs. Once a customer integrates a specific IP into its SoC architecture and aligns the software stack around it, moving to a different IP in the next chip generation carries redesign and re-verification costs. It’s not as sticky as changing foundries, but disturbing the whole pipeline just to swap one codec is unappealing. That inertia underwrites royalty recurrence.

But the moat is narrow. Here the decisive difference from ARM shows itself. ARM owns the CPU instruction set — a standard for an entire industry ecosystem, with software, developers, and toolchains stacked on top, so the cost of switching is astronomical. A video codec IP, by contrast, is closer to a well-defined functional block. A large SoC vendor that decides to can bring codec design in-house or substitute a competing IP or open-source implementation with relatively less friction. Chips&Media’s moat is not “a platform that owns a standard” but “stickiness resting on verification and switching friction.”

That distinction matters. Domain-specialist IP firms — Ceva (DSP and connectivity), Imagination (GPU IP), Verisilicon (NPU and platform IP) — each hold moats of a similar character in their niche: strong but narrow, and vulnerable to a single large customer’s in-house decision. Equate Chips&Media with an ARM-grade moat just because it licenses IP, and you’ll misjudge it.

---

On-Device AI and NPU: The Value and Limits of the Option

The most interesting piece of the bull case is the on-device AI option. AI compute migrating from the cloud onto the device itself is a structural opening for a codec IP company.

The logic runs like this. Smart cameras, autonomous and ADAS vehicles, CCTV and security devices, and AI appliances all need to process and analyze video inside the device in real time. That requires not just codecs (compress and restore) but image signal processing (ISP), super-resolution, and NPU compute for visual recognition. If Chips&Media can sell NPU and AI-acceleration IP and super-resolution features layered onto its existing codec blocks, both the breadth of IP adopted per customer and the royalty value per chip rise at once. Where it once sold a single codec, it now sells a bundle of video-processing IP.

If that plays out, two things change. First, higher royalty value means more revenue at the same shipment volume. Second, a front end that had leaned heavily on consumer electronics broadens into automotive, industrial, and security, diversifying cyclical exposure.

Stay clear-eyed, though. NPU and AI IP is not Chips&Media’s uncontested territory. It’s closer to a crowded field where Verisilicon, Ceva, and the in-house NPU teams of large SoC vendors are already fighting. On-device AI is a possibility, not booked earnings. New IP has to be designed into a customer’s chip, that chip has to reach volume production, and only then does it return as royalties through the long pipeline. How much premium to assign this option is the investor’s call, but never forget it is still a seed, not a harvest.

---

End-Market Cycles and Customer Risk: The Structural Small-Cap Weakness

Chips&Media’s royalties ultimately hinge on how well someone else’s chips sell. That dependency is the most frequently overlooked risk.

Consumer-electronics cycle. TVs, smart devices, and cameras — the front end where codec IP lands heavily — carry a strong discretionary character. When the economy slows, set-makers ship fewer units, the chips above them ship less, and the royalties above those shrink. Royalties sit at the end of a triple-linked chain, so front-end demand weakness reaches earnings with a lag.

Automotive and security-camera cycles. Automotive is a growth story but tied to vehicle production and inventory cycles, and security and CCTV are sensitive to infrastructure spending and geopolitics. Front-end diversification is a defense, but each front end brings its own cycle.

Customer in-house design and IP defection. The most structural risk. If a large customer decides to bring codec functionality in-house, the royalty base from that customer simply disappears. For a small-cap with high revenue concentration, that can be devastating. The company cannot control this risk, and it usually surfaces only after the fact.

Risk	Mechanism	Path to earnings
Consumer-electronics slowdown	Set and chip shipments fall	Royalty run-rate declines
Large customer goes in-house	Codec designed internally	A customer’s royalties vanish
Competing IP or open source	IP swapped in new chips	New royalty inflow slows
Small-cap liquidity and flows	Thin daily turnover	Amplified price volatility
Delayed new-IP adoption	AI and NPU pipeline slips	Growth expectations deferred

The table’s message is plain. Most of Chips&Media’s risks lie beyond “does the company execute well” — they sit in the front end, the customers, and the trading flows. You can own great IP and still watch royalties fall when the end market turns.

---

Chips&Media vs. the IP Licensing Peers: Where It Stands

Placing Chips&Media on the map of IP-licensing peers sharpens its character.

Company	Domain	Scale and moat width	Character
ARM	CPU instruction set and platform	Very large, industry-standard moat	Ecosystem lock-in, general purpose
Ceva	DSP, connectivity, AI	Mid-size, domain specialist	Wireless and edge IP breadth
Imagination	GPU, AI	Mid-size, domain specialist	Longtime graphics IP force
Verisilicon	NPU, platform IP	Mid-to-large, China-based	Design services plus IP
Chips&Media	Video codec, ISP	Small, narrow specialist	Codec focus, AI option

The key message of this comparison is that the same model does not mean the same tier. All five are in the same boat in that they take royalties on IP without a fab. But ARM’s moat defends an entire industry; Chips&Media’s defends one block called video codec. Investors should not equate Chips&Media with the majors under an “IP licensing equals premium multiple” frame — the accurate bucket is “a small IP name with a narrow, specialized, high-growth option.”

---

Three Practical Scenarios for the Individual Investor

Scenario 1: Korean Small-Cap Tax and Trading Strategy

Chips&Media is a domestic stock listed on KOSDAQ, and that changes the tax picture decisively versus a US name. For a retail investor below the large-shareholder threshold, capital gains on listed Korean shares realized on-exchange are tax-exempt — a contrast with an overseas stock like a US-listed ALGN, where gains are taxed at 22% including local surtax. On tax alone, the domestic small-cap has an edge.

Mind the large-shareholder threshold, though. Holding a single name above a certain ownership percentage or valuation can reclassify you as a large shareholder subject to capital-gains tax. If you concentrate a sizable sum in one small-cap, get in the habit of checking your holding size around the year-end record date.

👉 For the bigger picture on capital-gains reporting and tax efficiency, keep our overseas stock capital gains tax guide alongside as reference.

Scenario 2: Managing Liquidity and Volatility

The reality of a KOSDAQ small-cap is modest daily turnover. Good results can spike the stock in a flash; when flows leave, it can retrace in days. Try to move a large block the way you would in a large-cap and the fill itself works against you.

Three practical responses. First, size the position to what you can stomach — an individual small-cap should be a minor portfolio weight. Second, scale in and out to manage average cost and execution risk. Third, treat the wild swings around earnings and flow events as noise and keep your focus on the royalty run-rate as the underlying signal. The better the story, the more overheated the expectations tend to get on a small IP name, so the rule is to use only capital that can tolerate the volatility.

Scenario 3: Riding the IP Royalty Cycle

Chips&Media’s earnings ride a royalty cycle. The phase where new licenses accumulate (sowing seeds) and the phase where those licenses are harvested as royalties (fruit) are separated by a lag. The practical strategy is to exploit that lag.

A phase where license deals visibly increase and new customers and new IP (AI and NPU) adoption are confirmed is a reservation for royalties years out. Conversely, a phase where front-end chip shipments slow and the royalty run-rate rolls over is when the earnings leverage spins in reverse. Chase “royalties are great right now” late and you’re buying the cycle top; the “licenses are stacking but royalties haven’t arrived yet” phase may actually be the patient accumulation window. This judgment has to be made on leading indicators, not on hindsight — and the metrics in the next section are the tools.

---

Metrics to Watch Each Quarter

When you track Chips&Media, deciding in advance what to read first in the quarterly results makes judgment far cleaner.

First: the license-versus-royalty mix. Read both the weighting and the direction of the two revenue lines. A steadily rising royalty share signals that licenses sown earlier are bearing fruit and earnings leverage is showing up. Royalties stalling or falling should raise suspicion of end-market shipments or customer defection. Licenses can be lumpy; the trend in royalties reveals the underlying strength of the business.

Second: new licensing deals and customer diversification. Watch the count of new deals, the inflow of new customers, and any easing of concentration in a single large customer. The more diversified the customer base, the more the risk of any one project failing is spread. New licenses are a leading indicator of royalties years out — effectively the backlog of future growth.

Third: the direction of the royalty run-rate. Whether shipment-based recurring revenue is rising, flat, or rolling over is the core read. Because this ties directly to end-market chip shipment health, cross-check it against the consumer-electronics, automotive, and security front ends.

Fourth: new adoption of on-device AI and NPU IP. Look for concrete signals that AI, super-resolution, and NPU IP have been designed into actual customer chips. Progress here means both growth axes — higher royalty value and front-end expansion — are alive. If it stays perpetually an “option,” the case for a growth premium weakens.

Read those four together and you get past the “revenue grew X percent” headline to whether the royalty engine is actually accelerating or decelerating.

---

Related Reading

• 👉 ENF Technology (102710) Stock Outlook 2026: Semiconductor Process Materials
• 👉 AI Stocks Investment Guide 2026: Core Holdings and ETF Strategy
• 👉 Overseas Stock Capital Gains Tax Guide: Strategy and Practice
• 👉 SCHD Dividend ETF Guide 2026: Building an Income-Centered Portfolio

---

This article is for informational purposes only and does not constitute a recommendation to buy or sell any security. Investing in stocks involves risk, including possible loss of principal. All analysis reflects the author’s view as of the writing date; verify with current filings and consult a licensed financial professional before making investment decisions.