← back to rankingSPGI · S&P Global Inc.
Financial - Data & Stock Exchanges · mkt cap $123.6B · calls: Q1 FY2026 vs Q4 FY2025
62.0 conviction · conf-adj 59
conf 3/10 partial
enthusiasm:27.0 · trend:8 · quantifies:5 · impact:0 · under_radar:5 · credibility:5 · business_impact:8 · disruption:-6 · commitment:6 · confirmation:4
Enthusiasm latest 9 / prev 8 (rising)
Management’s AI thesis moved from strategic positioning in Q4 FY2025 to measurable commercial traction in Q1 FY2026, with customer counts, usage growth, ACV uplift, and renewal pricing premiums disclosed. The story is credible because it is tied to S&P Global’s own data, workflow products, Kensho technology, and customer renewals rather than generic AI claims. Bottom-line impact is still earlier-stage, with management saying Frontier model benefits are “just beginning” and more material in 2027 and 2028.
GROUNDED NEXT-FY IMPACT vs CONSENSUS
Grounded on actual base — revenue $15.3B · net income $4.5B · net margin 29.2% · diluted EPS 14.66
These are next-fiscal-year annual uplift estimates, not next-quarter numbers.
Aggregate next-FY est. rev uplift: 0.0% · next-FY EPS uplift: 1.06% · vs analysts: inline · priced in: medium (model's call-read: high; verdict above is the hard-data one used for ranking) · confidence: 3/10
| Claim | Figure | Arithmetic | Next-FY Rev % | Next-FY EPS % |
|---|
CapIQ Pro AI feature engagement
engagement · soft | >1/3 of CapIQ Pro users | Engagement share, not a paying take-rate; no per-user AI price or CapIQ revenue base disclosed -> cannot convert to $. | | |
AI applications customer count (~150)
engagement · soft | ~150 customers | Customer count across MI+Energy; no ACV per customer disclosed -> not sizeable. | | |
Kensho LLM-ready API customers (>300)
engagement · soft | >300 customers (contract or trial) | Mix of paid + trial, no per-contract price; pre-monetization. | | |
API call volume (>5x QoQ)
engagement · soft | >5x vs prior quarter | Usage metric; mgmt says economics not yet aligned to volume -> no revenue yet. | | |
API calls doubled Feb->Mar
engagement · soft | doubled MoM | Usage growth, unmonetized. | | |
MI ACV growth 30% higher for AI customers
revenue · soft | +30% relative growth | Relative growth uplift, but AI-customer share of MI revenue and baseline MI growth not disclosed -> cannot isolate $. | | |
Energy ACV growth 2x for AI customers
revenue · soft | 2x relative growth | No AI-customer revenue base or baseline rate disclosed. | | |
Kensho Labs client engagement (25%)
engagement · soft | 25% of CCO clients | Exploratory engagement, pre-revenue. | | |
CERA Titan demos (70 customers)
engagement · soft | 70 demo customers | Demo/pipeline metric, product hard-launches later this year; no revenue yet. | | |
Energy ChatAI queries >2x QoQ
engagement · soft | >2x queries | Usage metric, unmonetized. | | |
AI plug-in 20% pricing premium
revenue · soft | about 20% more expensive, still won | Single-client pricing-power anecdote; not a book-wide rate, affected revenue undisclosed. | | |
AI-ready renewal pricing uplift 35-45%
revenue · soft | 35% to 45% | Hard price-uplift rate but applied to an undisclosed share of the renewal book. Y illustrated ~1% rev/1.7% EPS assuming ~3% of recurring renews into AI, but that take is admitted-undisclosed; per the no-invent rule we leave it unsized. | | |
AI transformation resource coverage
other · soft | roughly half of resources | Scope statement framing the cost-out opportunity; no $. | | |
$1B AI investment since Kensho
other · soft | $1B cumulative | Cumulative spend (cost incurred), not an uplift; context for ROI only. | | |
iLEVEL automated ingestion adoption
revenue · soft | nearly 20% of iLEVEL customers | Paid add-on outside standard subscription -> genuinely incremental, but iLEVEL base and add-in price undisclosed -> cannot size. | | |
Automated data workflows (>half)
productivity · soft | >50% of workflows | Productivity driver underlying the EDO cost-out; no standalone $. | | |
Applications eliminated (>10%)
cost · soft | >10% of applications | Cost simplification feeding the EDO target; no standalone $ disclosed. | | |
With Intelligence Kensho Link data linking (>75%)
productivity · soft | >75% of datasets linked <1mo | Integration-speed efficiency on an acquisition; no $ disclosed. | | |
Research workflow savings
cost | $10M+ | $10M x (1-0.21 tax) = $7.9M after-tax; $7.9M / $4,471M NI = 0.18% EPS. Bottom-line only, topline ~0. | 0.0 | 0.18 |
Enterprise data office (EDO) cost reduction
cost | 20% of nearly $0.5B over 2 years | 0.20 x $500M = $100M over 2yr; next-FY straight-line ~half = $50M; $50M x 0.79 = $39.5M after-tax / $4,471M NI = 0.88% EPS. | 0.0 | 0.88 |
Assumptions: Net income base $4.471B, revenue $15.336B, current net margin ~29.15%. Tax rate 21% on cost savings. EDO savings phased straight-line, ~50% in next FY. Default incremental revenue margin would be the 29.15% net margin, but no revenue claim disclosed a base sufficient to size, so all revenue/engagement claims are soft and unmonetized. EPS-uplift figures expressed in percentage points.
Top line: Only two of eleven topline claims are even partially sizeable, and neither is dollar-anchored: management discloses engagement (>1/3 CapIQ users, >300 API customers, >5x call volume) and pricing power (35-45% renewal uplift for AI access, 20% premium plug-in that still won) but NO AI revenue figure. The cleanest hard rate (35-45% renewal uplift) lacks a disclosed share of book -> a conservative illustrative ~1% total-revenue uplift (~$153M, ~50% pure-price margin -> ~1.7% EPS), explicitly soft. Management itself says 'it will take some time to see exactly how this manifests in our financial results.'
Bottom line: The only dollar-anchored claims are cost-outs and they are small: research workflow $10M+ -> $7.9M after-tax = 0.18% EPS; EDO 20% of ~$0.5B = $100M over 2yr -> next-FY ~$50M -> $39.5M after-tax = 0.88% EPS (full run-rate 1.77%). Combined hard near-term EPS uplift ~1.06%, rising toward ~1.9% as EDO completes. Real but immaterial against a $4.47B earnings base.
Consensus already embeds FY26 revenue growth of (16,510-15,336)/15,336 = 7.66% and EPS growth of (19.62-17.86)/17.86 = 9.86%. The hard, dollar-quantified AI impact is ~1.1% EPS next-FY (cost) plus a soft ~1.7% from pricing flow-through = ~2.6% aggregate -> a slice of, and comfortably inside, the growth analysts already model. The large topline optionality (pricing power, API monetization) is undisclosed in dollars and management guides it to '27/'28 ('the upside... will have an impact in 27, 28 and future years'), so it is neither provable now nor plausibly baked into near-term consensus.
MODEL CONSENSUS (impact)
partial
Nearly all claims soft/unmonetized; only two hard cost figures size to ~1.06% EPS, immaterial vs consensus growth -> priced in, low confidence.
Conflicts reconciled
- renewal uplift eps_uplift_pct: X=null vs Y=1.7 -> used null because Y's ~3% renewal-take is admitted-undisclosed and the task forbids inventing a number (more conservative)
- research/EDO eps_uplift_pct units: X=0.0018/0.0088 (fractions) vs Y=0.18/0.88 (percent points) -> used percent points per the ~150% guardrail convention
- $1B investment soft: X=false vs Y=true -> used true because cumulative spend is not an uplift figure
| Field | Opus 4.8 | GPT-5.5 |
|---|
| Rev uplift % | 1.0 | 0 |
| EPS uplift % | 2.6 | 0.010601655110713487 |
| Priced in | medium | high |
| vs analysts | inline | behind |
| Confidence | 4 | 5 |
| Top line | Only two of eleven topline claims are even partially sizeable, and neither is dollar-anchored: management discloses engagement (>1/3 CapIQ users, >300 API customers, >5x call volume) and pricing power (35-45% renewal uplift for AI access, 20% premium plug-in that still won) but NO AI revenue figure. The cleanest hard rate (35-45% renewal uplift) lacks a disclosed share of book -> a conservative illustrative ~1% total-revenue uplift (~$153M, ~50% pure-price margin -> ~1.7% EPS), explicitly soft. Management itself says 'it will take some time to see exactly how this manifests in our financial results.' | Hard calculable next-FY revenue uplift is 0.0% because customer counts, engagement, usage growth, 20% premium, and 35%-45% renewal uplift lack affected revenue or ACV bases. |
| Bottom line | The only dollar-anchored claims are cost-outs and they are small: research workflow $10M+ -> $7.9M after-tax = 0.18% EPS; EDO 20% of ~$0.5B = $100M over 2yr -> next-FY ~$50M -> $39.5M after-tax = 0.88% EPS (full run-rate 1.77%). Combined hard near-term EPS uplift ~1.06%, rising toward ~1.9% as EDO completes. Real but immaterial against a $4.47B earnings base. | Hard quantified savings are $47.4M after tax: $7.9M from research workflow savings plus $39.5M next-FY EDO savings. That equals $47.4M / $4.471B net income = 1.06% EPS uplift, or about $0.16 per diluted share. |
| Reasoning | Consensus already embeds FY26 revenue growth of (16,510-15,336)/15,336 = 7.66% and EPS growth of (19.62-17.86)/17.86 = 9.86%. The hard, dollar-quantified AI impact is ~1.1% EPS next-FY (cost) plus a soft ~1.7% from pricing flow-through = ~2.6% aggregate -> a slice of, and comfortably inside, the growth analysts already model. The large topline optionality (pricing power, API monetization) is undisclosed in dollars and management guides it to '27/'28 ('the upside... will have an impact in 27, 28 and future years'), so it is neither provable now nor plausibly baked into near-term consensus. | Consensus FY2026 revenue is $16.510B, up 7.66% from the $15.336B base, while hard AI revenue uplift is 0.0%. Consensus FY2026 EPS is $19.623, up 33.86% from $14.66, versus hard AI EPS uplift of only 1.06%. The quantified AI math does not exceed the consensus trajectory; most potential upside is unquantified engagement and pricing optionality. |
Rows highlighted where the two models disagreed.
QUANTIFICATIONS
CapIQ Pro AI feature engagement: More than 1/3 of CapIQ Pro users (Q1 FY2026, topline)
“More than 1/3 of our CapIQ Pro users engage with the AI features we've launched, including ChatIQ and Document Intelligence.”
AI applications customer count: nearly 150 customers (March 2026 disclosure, topline)
“In March, we shared that nearly 150 customers across the Market Intelligence and Energy divisions, were interacting with our data through AI applications like Claude and Copilot.”
Kensho-LLM-ready API customers: more than 300 customers (Q1 FY2026, topline)
“We now have more than 300 customers under contract or in trial periods for Kensho-LLM-ready APIs.”
API call volume: more than 5x (Q1 FY2026 versus one quarter ago, topline)
“For instance, in the first quarter, the volume of API calls made by our customers was more than 5x the volume that we saw just 1 quarter ago.”
API call month-over-month growth: doubled (February to March 2026, topline)
“Volumes doubled month-over-month just from February to March.”
Market Intelligence ACV growth uplift from AI customers: 30% higher (Q1 FY2026, topline)
“Growth in Market Intelligence is 30% higher among AI customers compared to others and growth among AI customers and Energy is double the growth rate among other customers.”
Energy ACV growth uplift from AI customers: double the growth rate (Q1 FY2026, topline)
“Growth in Market Intelligence is 30% higher among AI customers compared to others and growth among AI customers and Energy is double the growth rate among other customers.”
Chief Client Office Kensho Labs engagement: 25% of these clients (Q1 FY2026, topline)
“25% of these clients are engaged with our Kensho Labs Technologies to explore opportunities to leverage our technology and data to help solve their most challenging problems.”
CERA Titan demos: 70 customers (Q1 FY2026, topline)
“70 customers were able to demo the new platform and feedback was overwhelmingly positive.”
Energy ChatAI usage: more than doubled (Q1 FY2026 quarter-over-quarter, topline)
“CERAWeek in Houston hit new records and online, the number of user queries in our Energy platforms, ChatAI feature more than doubled quarter-over-quarter.”
AI plug-in pricing premium: about 20% more expensive (Q1 FY2026, topline)
“And as a result, the clients liked it so much that they actually canceled their existing provider and went with our data and plug-in even though it was about 20% more expensive.”
AI-ready renewal pricing uplift: 35% to 45% (Q1 FY2026, topline)
“And were willing to pay in the range of 35% to 45% on the renewal increase to get the AI access.”
Strategic processes covered by AI transformation: roughly around half of the resources (Q1 FY2026, bottomline)
“And these comprise roughly around half of the resources that we have at the company.”
AI investment: $1 billion (since Kensho acquisition in 2018 through Q4 FY2025, both)
“We've deployed about $1 billion against this, and that's really put us in a great position as we think about deploying these capabilities, both within our products and in our internal processes.”
iLEVEL automated ingestion adoption: nearly 20% of the iLEVEL customers (within 6 months in 2025, topline)
“We deployed the automated data ingestion tool on iLEVEL in 2025. And with the 6 months, we had nearly 20% of the iLEVEL customers opting for that add in, which is not part of the standard subscription.”
Automated data workflows: more than half (2025, bottomline)
“In 2025 alone, we reduced manual data processing meaningfully with more than half of our total data workflows now processed via automation tools.”
Applications eliminated: more than 10% (2025, bottomline)
“We also eliminated more than 10% of applications in use and simplified the EDO technology stack to standardize on the best applications and reduce costs.”
With Intelligence data linking via Kensho Link: more than 75% (less than a month after close, both)
“We linked more than 75% of the fund manager and investor data sets in less than a month through the application of Kensho Link.”
Research workflow savings: $10 million plus (over the last year, bottomline)
“researchers in an area where we've already been able to simplify, streamline and save $10 million plus over the last year as we provide them the tools and the functionality and the capabilities that they need to provide more research faster and more efficiently than before.”
Enterprise data office cost reduction target: 20% reduction (over the next 2 years, bottomline)
“And then the effort that's probably the furthest ahead is the enterprise data office where over time, over the next 2 years, we see about a 20% reduction in that cost base in that area.”
Enterprise data office expense base: nearly a $0.5 billion expense base (Q4 FY2025, bottomline)
“It's nearly a $0.5 billion expense base out of $7.5 billion.”
PAST (realized)
- 2025 was truly a leap forward for S&P Global in AI.
- We launched new AI products and features in every division, many of which were on display at our Investor Day.
- In 2025 alone, we reduced manual data processing meaningfully with more than half of our total data workflows now processed via automation tools.
- We linked more than 75% of the fund manager and investor data sets in less than a month through the application of Kensho Link.
- We've deployed about $1 billion against this, and that's really put us in a great position as we think about deploying these capabilities, both within our products and in our internal processes.
CURRENT (now)
- More than 1/3 of our CapIQ Pro users engage with the AI features we've launched, including ChatIQ and Document Intelligence.
- We now have more than 300 customers under contract or in trial periods for Kensho-LLM-ready APIs.
- For instance, in the first quarter, the volume of API calls made by our customers was more than 5x the volume that we saw just 1 quarter ago.
- ACV growth among customers who use our AI solutions is outpacing growth from other customers by a wide margin.
- There's certainly a set of AI benefits that we're getting as we think about our data operations, which is a big part of MI.
- We see emerging progress or I think I'd say good progress in software development activities that are AI-driven with all the new tools available to it.
FORWARD (guidance)
- While it will take some time to see exactly how this manifests in our financial results, we are confident that the value we create for our customers is increasing and the economics will reflect that over time.
- And as usage increases and use cases expand, we expect to align the economics with the value we create through price.
- The official hard launch for the product is going to be a little bit later this year.
- We also saw significant debt issuance from hyperscaler investments in AI infrastructure in the second half of 2025, and we expect that to continue in 2026, albeit spread more throughout the year.
- I think the upside from the broad adoption of Frontier models is just beginning. And really will have an impact in '27, '28 and in the future years as they get expanded into a wide range of these strategic and important processes that we operate and will be helpful in that regard.
TRACK RECORD — PROMISE vs DELIVERY
65/100 track record too-early 6 calls reviewed
S&P Global rarely sets hard number-plus-date AI targets; its one clearly quantified commitment (>20% EDO cost reduction by 2027) is still too-early, while the softer adoption milestones it has touted — internal copilot usage, Kensho API customer ramp, and AI-assisted integration timelines — have been met or beaten, so the limited track record looks credible but not yet fully judgeable.
Cut enterprise data run-rate expenses >20% by end of 2027 via automation/AI tools (Enterprise Data Office) — promised Q4 FY2025
too-early Reported 'well ahead of pace' — already >50% of data workflows automated and >10% of apps eliminated in 2025, but 2027 deadline not yet reached
Spark Assist internal AI copilot adoption scaling across workforce (started ~30% / 14,000 users) — promised Q4 FY2024
delivered Grew to >65% of global workforce actively using it by Q2 FY2025 and published Sparks more than doubled to >3,000 — adoption milestone clearly hit
Kensho LLM-ready API customer adoption and data-consumption ramp — promised Q2 FY2025
partial By Q1 FY2026 >300 customers under contract/trial (from ~150 in March) and API call volume grew >5x quarter-over-quarter — trending strongly positive
AI-powered SPICE index builder cutting custom index creation from ~1 month to 2 days — promised Q2 FY2025
too-early Stated as an achieved capability at launch; no later quantified validation or recurring metric provided
With Intelligence data integration via Kensho Link on accelerated timeline — promised Q4 FY2025
delivered Linked >75% of fund-manager/investor datasets in under a month and closed deal in <6 weeks vs. original plan — beat the stated timeline
PRICED-IN (REFINED)
MEDIUMEst. revisions flat · Fwd P/E 23.4 · EV/Sales 8.6x
AI claim maps to Market Intelligence Segment, Ratings Segment, Commodity Insights
Analyst ratings have been essentially flat in recent months, with no usable recent price-target momentum because last-month and last-quarter averages are zero. Forward revenue and EPS estimates show steady growth, but not a sharp upward revision signal. Valuation is rich at 23.4x forward earnings and 8.6x TTM EV/sales, so AI upside tied mainly to Market Intelligence, Ratings, and Commodity Insights is partly reflected despite flat revisions. Rich valuation with flat estimate momentum points to a medium priced-in verdict, not high because estimates are not clearly rising.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
6Q4 FY20242Q1 FY20257Q2 FY20258Q3 FY20259Q4 FY202510Q1 FY2026
AI enthusiasm across 6 calls — trend ↗ rising
AI moved from broad investment theme to concrete products, partner distribution, customer adoption metrics, API usage growth, and ACV uplift.
RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY
8/10 qualitative impact material medium-term · mixed evidence
Where AI matters: Market Intelligence, Energy data APIs, workflow automation
AI is already embedded in core S&P Global data/workflow products, with >1/3 of CapIQ Pro users engaging AI features, >300 Kensho LLM-ready API customers/trials, >5x API-call growth, and higher ACV growth among AI customers. This is material because it improves product utility and pricing power in key subscription franchises, but company-level Rev/EPS uplift is still not sized and management frames larger benefits as 2027-2028.
Caveats: AI usage growth may not convert into incremental paid ACV or pricing power; Platform partners and large AI vendors could capture distribution economics; Agentic workflows may cannibalize traditional seats or reduce demand for lower-value research interfaces; Regulatory, data-rights, hallucination, and liability issues could slow adoption in financial workflows
AI DISRUPTION / CANNIBALIZATION RISK two-sided · 4/10
AI could pressure seat-based research terminals, basic data interfaces, and lower-value analyst workflow content as customers consume answers through agents instead of traditional products. The core model is still fairly durable because S&P owns scarce proprietary datasets, ratings/index franchises, benchmarks, and licensed workflow data that AI systems need rather than easily replace.
OPTIONS / MARKET STRUCTURE
option liquidity: good
proxy inputs — dollar-ADV $848M · beta 1.105 · px $417.46
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.
CONFIRMATION — INSIDERS · 13F · LANGUAGE
Confirming — insiders buying, institutions flat, management language 7/10 committed.
INSIDERS buying 6 open-market buy(s) vs 0 sell(s) — net accumulation
INSTITUTIONS (13F) flat as of 2026-03-31: 213 new / 308 closed positions; 975 increased / 866 reduced; institutional ownership -2.88pp; -103 net 13F holders
MGMT LANGUAGE 7/10 committed Concrete adoption, usage and ACV metrics show ownership, though monetization timing remains qualified and partly exploratory.
commit “More than 1/3 of our CapIQ Pro users engage with the AI features we've launched, including ChatIQ and Document Intelligence.”
commit “We now have more than 300 customers under contract or in trial periods for Kensho-LLM-ready APIs.”
commit “ACV growth among customers who use our AI solutions is outpacing growth from other customers by a wide margin.”
VERBATIM AI QUOTES
“We are also advancing our leadership as we help our customers unlock the potential of AI.”
— Martina Cheung, Q1 FY2026
“More than 1/3 of our CapIQ Pro users engage with the AI features we've launched, including ChatIQ and Document Intelligence.”
— Martina Cheung, Q1 FY2026
“We now have more than 300 customers under contract or in trial periods for Kensho-LLM-ready APIs.”
— Martina Cheung, Q1 FY2026
“For instance, in the first quarter, the volume of API calls made by our customers was more than 5x the volume that we saw just 1 quarter ago.”
— Martina Cheung, Q1 FY2026
“ACV growth among customers who use our AI solutions is outpacing growth from other customers by a wide margin.”
— Martina Cheung, Q1 FY2026
“Growth in Market Intelligence is 30% higher among AI customers compared to others and growth among AI customers and Energy is double the growth rate among other customers.”
— Martina Cheung, Q1 FY2026
“We unveiled our new AI native Upstream product for data and insights called CERA Titan.”
— Martina Cheung, Q1 FY2026
“Energy & Resources, Data & Insights and Price Assessments grew 7% and 6%, respectively, driven by strength in petroleum gas, power and renewables.”
— Eric Aboaf, Q1 FY2026
“CERAWeek in Houston hit new records and online, the number of user queries in our Energy platforms, ChatAI feature more than doubled quarter-over-quarter.”
— Eric Aboaf, Q1 FY2026
“There's certainly a set of AI benefits that we're getting as we think about our data operations, which is a big part of MI.”
— Eric Aboaf, Q1 FY2026
“we can see tremendous capacity expansion within Ratings, for example, where they have been a very early adopter of AI as part of augmenting analytical capacity and making sure that our analysts can do more high-value things like thought leadership and additional research.”
— Martina Cheung, Q1 FY2026
“2025 was truly a leap forward for S&P Global in AI.”
— Martina Cheung, Q4 FY2025
“We launched new AI products and features in every division, many of which were on display at our Investor Day.”
— Martina Cheung, Q4 FY2025
“We saw significant debt issuance from hyperscaler investments in AI infrastructure in the second half of 2025, and we expect that to continue in 2026, albeit spread more throughout the year.”
— Martina Cheung, Q4 FY2025
“In 2025 alone, we reduced manual data processing meaningfully with more than half of our total data workflows now processed via automation tools.”
— Martina Cheung, Q4 FY2025
“We linked more than 75% of the fund manager and investor data sets in less than a month through the application of Kensho Link.”
— Martina Cheung, Q4 FY2025
“And we're pairing these tools with end-to-end process reengineering, so the productivity gains are sustainable, long-term value generators that scale, not just isolated use cases.”
— Eric Aboaf, Q4 FY2025
“We deployed the automated data ingestion tool on iLEVEL in 2025. And with the 6 months, we had nearly 20% of the iLEVEL customers opting for that add in, which is not part of the standard subscription.”
— Martina Cheung, Q4 FY2025
“researchers in an area where we've already been able to simplify, streamline and save $10 million plus over the last year as we provide them the tools and the functionality and the capabilities that they need to provide more research faster and more efficiently than before.”
— Eric Aboaf, Q4 FY2025
ANALYST QUESTIONS ON AI
Q (Q1 FY2026, Toni Kaplan): Martina, thanks for the color on what you're doing with regard to the AI distribution channels. I was hoping that you could expand on how you're thinking about the partnership strategy with the large AI players? Are you building SMP, MCP apps on the platforms? Or you just plan to continue to provide the data through the MCP integrations and the APIs? And maybe if you could just talk about the monetization model and directional economics between the different distribution channels.
A: The first step to doing that has actually been the announcement of the S&P Global plug-in, which was announced in line with the Claude for Financial Services announcement in the first quarter. And that's essentially a series of agents that teach AI agents within the platform, how to actually conduct specific tasks for data, AI-ready data that the client might be licensed to. So maybe to give you an example, one of our buy-side clients working with Kensho was looking at our financial data via at AI-ready API and Kensho helped them to understand how to use the plug-in to perform tasks like creating tearsheets or creating earnings calls previews. And as a result, the clients liked it so much that they actually canceled their existing provider and went with our data and plug-in even though it was about 20% more expensive.
Q (Q1 FY2026, Scott Wurtzel): On the Market Intelligence margin, just wondering if you can maybe help contextualize how much of the margin expansion that you're seeing is being driven by efficiency gains associated with AI.
A: There's certainly a set of AI benefits that we're getting as we think about our data operations, which is a big part of MI. We see emerging progress or I think I'd say good progress in software development activities that are AI-driven with all the new tools available to it.
Q (Q1 FY2026, Craig Huber): I wanted to ask about AI efficiencies at your company. To the extent that you can give us some more examples of how AI internally is helping you guys be more efficient across your various sectors, including outside of the MI division? And also, Eric, I wanted to ask your 50 to 75 basis points expected improvement, excluding OSTTRA, how much ballpark do you think AI efficiencies is actually helping that number?
A: I would say that we have been tackling AI by looking at some of our largest strategic processes across the company. And so at our IR Day, for example, we mentioned four particular areas that we were focused on, including our Ratings analytic workflows, our research workflows in Energy and in Market Intelligence as well as our technology and data workflows. And these comprise roughly around half of the resources that we have at the company.
Q (Q1 FY2026, Owen Lau): So following up on the AI Upstream data platform Titan, it's still in beta testing version. But could you please talk about your go-to-market strategy and the revenue model of this product? Is it going to be a subscription-based model or consumption-based or a combination of the two?
A: It's going to be a subscription-based model. And in terms of the broader go-to-market strategy, I think the team was able to really effectively leverage CERAWeek because we have so many clients in town to be able to do our launch and get this into the minds of so many of our customers.
Q (Q4 FY2025, Keen Fai Tong): Anthropic recently announced a suite of 11 open source plug-ins for Claude cohort. Can you talk a bit about how you expect this competitive development to impact S&P's business?
A: Look, we think these kinds of announcements are really exciting. And we're actively involved in advancing this technology and actually helping to establish these ecosystem ourselves. As you know, we've worked with pretty much every major player in the AI space for some time. And we see AI really is a net tailwind for the business.
Q (Q4 FY2025, Faiza Alwy): So I wanted to follow up on George's question, just given the panic and confusion that the market is experiencing as it relates to AI. I know you've talked previously about revenues driven by benchmarks and proprietary data. And I think you've said you're fairly agnostic around the channel that data consumption occurs in. So I was hoping you could put a finer point on that and talk specifically about your workflow products and the mood there.
A: But the workflow tools that S&P Global has developed are critical systems of record for our customers. So products like iLEVEL, ClearPar Cap IQ Pro, Platts Connect and others, they're not simple apps that were developed rapidly. And in fact, they get smarter as we embed AI technology in them.
Q (Q4 FY2025, Surinder Thind): At a high level, can you maybe talk about your assessment and experience with the AI technology in your attempts to deploy it internally versus maybe the hype that's coming out of Silicon Valley?
A: We've deployed about $1 billion against this, and that's really put us in a great position as we think about deploying these capabilities, both within our products and in our internal processes.
Q (Q4 FY2025, Manav Patnaik): Martina, you talked about how you thought AI was net tailwind for your business. And I was hoping you could just elaborate on that more on the top line basis. So do you think all these enhancements that you're talking about will help you accelerate revenue growth? And also, how do you think that changes your pricing strategy going forward?
A: First, I'd say that our clients are getting additional value by being able to use the data, our data in more ways. And the more ways to use it, the more value it creates and the better opportunity for value-based conversation at renewal when we talk to those customers.