conf 5/10 partial
enthusiasm:9.0 · trend:0 · quantifies:5 · impact:0 · under_radar:5 · credibility:0 · business_impact:4 · disruption:-14 · commitment:-4 · confirmation:3
Enthusiasm latest 3 / prev 4 (flat)
Across both calls, management never names AI; the substantive story is data-driven personalization, search/recommendation merchandising in-car, trial conversion marketing, and ad targeting/measurement (360L, YouTube, programmatic)—not generative AI or a distinct AI product. Credibility is modest: they cite engagement lift from personalized 360L listening and small-scale conversion tests, but in-car targeted ads and scaled personalized marketing are still largely forward; no figure attributes revenue, margin, or savings specifically to AI/ML.
Grounded on actual base — revenue $8.6B · net income $0.8B · net margin 9.4% · diluted EPS 2.25
These are next-fiscal-year annual uplift estimates, not next-quarter numbers.
Aggregate next-FY est. rev uplift: % · next-FY EPS uplift: 3.96% · vs analysts: inline · priced in: medium (model's call-read: high; verdict above is the hard-data one used for ranking) · confidence: 5/10
| Claim | Figure | Arithmetic | Next-FY Rev % | Next-FY EPS % |
|---|---|---|---|---|
| 360L personalized listening — double-digit usage/time growth engagement · soft | double-digit growth in usage & time spent (Q1 FY2026) | Engagement metric only — no 360L revenue, ARPU, or $ base disclosed in any claim; 'double-digit' applies to usage/time, not revenue. Directionally supports retention/ad-inventory but unsizable from inputs → rev_uplift_pct and eps_uplift_pct null. | ||
| Programmatic ad revenue via Google TV 360 more than doubled YoY revenue · soft | >100% YoY (Q1 FY2026) | Growth rate with NO disclosed base anywhere in claims (programmatic/Google TV 360 ad-revenue dollars not given) → cannot compute incr_rev or rev_uplift_pct=100×incr/$8,558M. Doubling of an unknown small ad slice; positive but unsizable. | ||
| Podcast programmatic demand — triple-digit growth YoY revenue · soft | triple-digit % YoY (Q1 FY2026) | ≥100% YoY 'demand' (bookings-like, not recognized revenue) with no disclosed $ base in claims → no incr_$ → null rev/EPS. | ||
| Podcast programmatic demand +92% over Q4 2024 revenue · soft | >92% over Q4 2024 (Q4 FY2025) | Same demand metric, +92% QoQ with no Q4 2024 $ base disclosed → cannot anchor incr_rev. Unsizable. | ||
| $100M additional gross cost savings exiting 2026 ($350M cumulative run-rate) cost | +$100M gross (exiting 2026); $350M cumulative | DISCLOSED $100M incremental gross (use new layer only; $350M cumulative largely in FY25 base). It is a run-rate EXITING 2026, so NEXT-FY (FY2026) realization is partial: assume ~50% average → $50M gross in P&L → after-tax @21% $50M×0.79=$39.5M. EPS vs consensus adjusted NI: 100×$39.5M/$997,345,547=3.96%. Both analysts agree full $100M (→$79M after-tax, 7.9%) is a FY2027 event. rev_uplift_pct null (cost, not topline; mis-stated as rev=100×$50M/$8,558M=0.58%). | 3.96 |
Assumptions: Adopter-side only. Tax 21% on gross cost savings (dollar-for-dollar opex; incremental net margin not applied). Cost claim uses only the NEW $100M layer (not $350M cumulative, to avoid double-count with prior savings already in FY25 base). FY2026 phasing: $100M is run-rate EXITING 2026, so assume ~50% average in-year realization ($50M gross → $39.5M after-tax); full $100M lands FY2027 — Y's own '~$40-50M after-tax in 2026' corroborates this. EPS denominator: consensus adjusted NI FY2025 ($997.3M / $2.79 EPS), matching non-GAAP basis; GAAP FY2025 NI $805M / EPS $2.25 is depressed and would overstate the % by ~24%, not used. Net margin 9.4% (>3%), no thin-denominator artifact. AI/CS automation is only ONE of three named drivers (with platform efficiency + G&A), so 3.96% is an upper bound for the whole program, not AI-only. Engagement/programmatic claims: growth rates with no segment $ → soft. Forward AI ad/measurement statements unquantified, excluded.
Top line: Not sizable from disclosed data. Every revenue/engagement AI claim — 360L double-digit engagement, programmatic ad revenue >2× YoY, podcast programmatic triple-digit YoY and +92% QoQ — is a growth RATE with no dollar base disclosed in any quote. Real and directionally positive (doubling a small programmatic slice; deeper 360L engagement feeding ad inventory and churn) but unsizable without inventing the base. Consensus revenue is essentially flat (FY2026 $8.566B vs FY2025 $8.537B, +0.34%), so the Street is NOT modeling material AI-driven topline. est_rev_uplift_pct=null.
Bottom line: Only anchored hard impact is bottom-line: the new $100M gross cost-savings layer exiting 2026. NEXT-FY (FY2026) P&L realization is partial (~50% → $50M gross → $39.5M after-tax = 3.96% of adjusted NI $997.3M, ~0.58% of revenue). Full $100M run rate (→$79M after-tax, 7.9%) is a FY2027 phenomenon. AI/CS automation is one of three drivers, so even the phased figure is an upper bound for the AI-attributable slice. No supplier-side EPS.
Consensus FY2026 shows NI $997M→$1,134M (+13.7%) and EPS $2.79→$3.13 (+12.3%) on essentially FLAT revenue (+0.34%). EPS growth with no revenue growth IS margin expansion — precisely the cost-savings program, which is management-guided and thus already embedded. The phased $39.5M after-tax explains ~28.9% of the $137M consensus NI step-up (full $100M run rate ~57.7%), confirming efficiency/margin is in the street ramp while revenue is flat. The unsized revenue claims aren't in consensus either (flat top line), so no clear above-consensus gap. FY2027 consensus adds only +1.29% revenue and +12.4% NI — also consistent with cost-led EPS, not AI topline. Unpriced optionality (targeted ads in 360L, fall 2026 ad-targeting/measurement) is unquantified by management.
partial
Agree all topline claims unsizable (no $ base) and impact is priced-in/inline. Conflict only on cost-saving phasing; took X's next-FY phased 3.96%, which Y's own caveat supports.
| Field | Opus 4.8 | GPT-5.5 |
|---|---|---|
| Rev uplift % | – | – |
| EPS uplift % | 7.9 | – |
| Priced in | high | – |
| vs analysts | inline | – |
| Confidence | 6 | – |
| Top line | Not sizable from disclosed data. Every revenue/engagement AI claim — 360L double-digit engagement, programmatic ad revenue 'more than doubled,' podcast programmatic 'triple-digit'/+92% — is a growth RATE with no dollar base disclosed in any quote. These are real and directionally positive (doubling a small programmatic slice, deeper 360L engagement feeding ad inventory and churn), but unsizable: a 2x on an undisclosed base cannot be translated to a credible revenue % without inventing the base. Consensus revenue is essentially flat (+0.3% 2025→2026), so the Street is NOT modeling material AI-driven topline. est_rev_uplift_pct = null. | – |
| Bottom line | The only anchored, hard impact is bottom-line: the new $100M gross cost-savings layer exiting 2026 → ~$79M after-tax (21%) = ~7.9% of adjusted net income ($997M), ~1.2% of revenue. But (a) it's a run-rate exiting the year, so 2026 P&L realization is partial (~$40-50M after-tax), full benefit is a 2027 phenomenon, and (b) AI/customer-service automation is only ONE of three named drivers (with platform efficiency and G&A), so 7.9% is an upper bound for the entire program, not the AI-attributable portion alone. | – |
| Reasoning | Consensus 2026 shows net income $997M→$1,134M (+13.7%) and EPS $2.79→$3.13 (+12.3%) on essentially FLAT revenue (+0.3%). EPS growth with no revenue growth IS margin expansion — precisely the cost-savings program. The $79M after-tax AI/efficiency layer (~7.9% of NI) fits inside the $137M consensus NI step-up and is explicitly management-guided, so it is already embedded. The unsized revenue claims aren't in consensus either (flat top line), so there's no clear above-consensus gap to exploit. Net: anchored impact is real but already priced; the unpriced optionality (targeted ads in 360L, fall 2026 ad-targeting/measurement) is unquantified by management. | – |
Rows highlighted where the two models disagreed.
—/100 (no quantified promises) no-quantified-promises 6 calls reviewed
Across six calls (Q4 FY2024–Q1 FY2026), SiriusXM management referenced AI only qualitatively (customer experience, marketing, ads, AI voice replicas for creative, AI-enhanced development) and reported operational metrics (personalization, 360L penetration, podcast/programmatic growth) without ever stating a numbered AI outcome with a deadline. Cost-savings ($200M by end-2025, $100M in 2026) and financial targets ($1.5B FCF by 2027) were quantified but not framed as AI deliverables, so AI promise-vs-delivery credibility cannot be scored from this transcript set.
Est. revisions rising · Fwd P/E 10.3 · EV/Sales 2.3x
AI claim maps to Subscription and Circulation, Advertising, Other Revenue
AI enthusiasm across 6 calls — trend ↘ falling
No AI story in late 2024; brief generic AI in Q1 FY25, one concrete ads voice-replica launch in Q2, then mostly data/personalization without explicit AI.
5/10 qualitative impact moderate medium-term · mixed evidence
Where AI matters: 360L personalization, ad targeting, cost automation
Deployed ML-style personalization and search/recommendation are lifting 360L engagement and supporting programmatic ad growth, but management never ties dollars to AI and consensus revenue is flat; the only hard P&L anchor is a partial slice of $100M gross savings exiting 2026 where CS automation is one of three drivers.
Caveats: Management rarely names AI; upside is classical data/ML framed as personalization; In-car targeted 360L ads and scaled personalized marketing are still forward, not proven revenue; $100M savings and ~4% EPS uplift are partly non-AI and largely already in margin-led consensus; Programmatic/podcast growth rates lack disclosed bases—may be immaterial to ~$8.5B revenue
AI DISRUPTION / CANNIBALIZATION RISK two-sided · 6/10
GenAI and AI-native streaming raise the bar for discovery, lean-back playlists, and cheap audio supply, pressuring commoditized listening and ad yield, while SiriusXM's moat rests on exclusive licenses, OEM/car distribution, and sports/talk—not on being an AI platform.
option liquidity: fair
proxy inputs — dollar-ADV $136M · beta 0.943 · px $28.61
source: proxy (no options chain on FMP)
FMP /stable/ exposes no options-chain endpoint on this key, so ATM IV, bid-ask spread and open interest are unavailable. Liquidity below is a PROXY from dollar-ADV, beta and price level (a stand-in for option depth), not measured option-market data.