AIP · Arteris, Inc. · Stock research

Current Thesis

The business inflected further and the stock kept falling. Q2 2026, reported 2026-08-06, put revenue at $24.1M (+46% YoY) against a ~$23.5M consensus, with trailing-twelve-month variable royalties — the line the entire thesis rests on — at a record $8.6M, +65% YoY. ACV plus royalties reached $99.5M (+44%), above the top of guidance, and RPO hit $135M (+36%). Management raised the FY26 revenue guide to $95–98M from $91–95M and set Q3 at $24–25M versus a ~$23.4M estimate.

The tape rejected all of it. Shares slid ~5.8% to $29.78 on 2026-08-08 and closed at $28.32 on 2026-08-14 — 41.7% below the $48.59 closing high of 2026-06-30 and below the ~$32–34 July reaction low.

That resolves the two conditions the July note carried. The price condition (a weekly close below $33) fired; the fundamental condition (variable-royalty growth decelerating under +25% YoY) did not — it printed +65%. The narrative is dead. The re-rating narrative that carried the name from ~$19.56 in early April to $48.59 on 6/30 failed at the June top, and the structure that supported it — the ~$39 breakout shelf, then the July low — is gone. A beat-and-raise sold into is the market marking the multiple down while estimates go up.

Bullish and bearish views on Arteris, Inc.

The model's bull view on Arteris, Inc. (AIP), in brief: Q2 2026 (2026-08-06): revenue $24.1M, +46% YoY, above the ~$23.5M consensus and above the $23–24M guide — an acceleration from Q1's +39%. The bear view: Losses widened while revenue accelerated. Both cases follow in full.

Bull Case

• Q2 2026 (2026-08-06): revenue $24.1M, +46% YoY, above the ~$23.5M consensus and above the $23–24M guide — an acceleration from Q1's +39%.
• TTM variable royalties $8.6M, +65% YoY, a company record. The KPI cleared the +25% floor flagged as thesis-breaking by roughly 40 percentage points.
• ACV plus royalties $99.5M (+44%), above the guided range; RPO $135M (+36%), an all-time high — contracted visibility grew faster than reported revenue.
• FY26 guide raised twice in 2026: $91–95M on 2026-05-12, then $95–98M on 2026-08-06, with FY26 ACV plus royalties $102–106M and free cash flow guided $5–9M.
• Design starts +21% over the trailing twelve months, and CEO K. Charles Janac said on 2026-08-06 that enterprise computing now averages 29% of ACV plus royalties — mix migrating from ADAS toward AI datacenter silicon.
• Balance sheet at 2026-06-30: $93.3M cash and equivalents plus $28.8M short-term investments against a guided FY26 non-GAAP operating loss of $7–10M.
• Sell-side moved up after the print: Jefferies upgraded to Buy, target $50 (2026-08-08); Roth Capital initiated Buy, target $40 (2026-08-04).

Bear Case

• Losses widened while revenue accelerated. Q2 GAAP net loss was $0.30/share; non-GAAP $0.10 versus a $(0.04) consensus. A 46% revenue quarter did not narrow the loss line.
• The print was the test and it failed on price. No higher close has printed since 2026-08-06; $28.32 on 2026-08-14 sits under every level this cycle was built on.
• Structure is exhausted below spot. The ~$39 shelf broke in early July, the ~$32–34 reaction low broke after the print. The next reference inside 2026's range is the early-April area near $19.56.
• Targets have not been marked to price. Published targets run $35 low to $50 high, with aggregator averages of $38.25 (WallStreetZen, 4 analysts) and $41.00 (6 analysts) as of mid-August — every published number sits above spot, so downward revisions remain available.
• CFO turnover lands inside the de-rating. Nicholas Hawkins retires 2026-08-31 (announced 2026-05-08, family illness); Saurabh Sinha starts 2026-09-08 (announced 2026-08-06). The first full guidance cycle under new finance leadership is the Q3 report.
• Royalties lag design wins by roughly 18–24 months, so +65% reflects 2024–25 bookings. Softness in current design starts would not surface in the royalty line until 2027.

Setup & Price Structure

• Reference close $28.32 (2026-08-14). 52-week high $48.59. Distance from high −41.7%. Three-month return −16.0%. RSI(14) 44.3.
• MEASURED: RSI in the mid-40s after a 41.7% drawdown. INFERRED: the decline has been a sequence of lower highs rather than one capitulation flush, so there is no washed-out momentum reading to lean on.
• The path since the top: $48.59 close (2026-06-30) → loss of the ~$39 shelf (early July) → ~$32–34 reaction low → $29.78 (2026-08-08) → $28.32 (2026-08-14). Six weeks of lower highs and lower lows; no base has formed.
• First evidence of repair would be a weekly close back above the ~$34 shelf lost after the print. Until that exists, rallies are counter-trend inside a broken structure.
• Positioning observables: no scheduled earnings inside 30 days, so no imminent binary; the entire published target range sits above spot; float is small (~44M shares), which keeps 5–8% single sessions routine without company news.

Catalyst Calendar (next 30 days)

• 2026-08-31 — Nicholas Hawkins's retirement as CFO takes effect (announced 2026-05-08).
• 2026-09-08 — Saurabh Sinha begins as CFO (announced 2026-08-06). Closes the vacancy; his first guidance cycle is the Q3 report.
• ~2026-11-04 (est.) — Q3 2026 print, extrapolated from the 2026-08-06 Q2 cadence. Outside the 30-day window; it is the next event that can move estimates.

What Would Change Our Mind

The structural repair evidence is a weekly close back above $34, reclaiming the shelf that broke after the 2026-08-06 print, followed by the Q3 report holding TTM variable-royalty growth above +40% YoY and the FY26 revenue guide at or above $95–98M. That combination would argue the August de-rating was multiple compression on an intact royalty ramp.

The break condition is a weekly close below $26, which extends the post-print de-rating to fresh lows and confirms the beat-and-raise drew no bid at any price on the way down.

The fundamental flip: TTM variable royalties growing under +25% YoY at the Q3 print, or an FY26 revenue guide cut back toward $91–95M, removes the one leg that survived August.

The catalyst that comes and goes: if the 2026-09-08 CFO start passes and the stock still cannot hold $30 into month-end, the leadership overhang was not what the market was pricing.

Correlation Notes

• Levered to the large-cap AI-semi complex (NVDA/AVGO/TSMC/ARM/Marvell). It amplifies that factor rather than diversifying it, and the small float makes both directions larger.
• Arm Holdings is the closest listed comparison on royalty-model mechanics — per-unit royalties trailing licensing by years — so Arm's royalty commentary reads across.
• Automotive design wins (Li Auto, SiEngine, both cited 2026-08-06) tie part of the royalty base to China EV production volumes.
• With enterprise computing at 29% of ACV plus royalties, hyperscaler custom-silicon programs are becoming a larger swing factor than ADAS; hyperscaler capex commentary now matters more to the royalty line than auto build rates.