Ideas
AI apps driving net new software revenue.
The AI era is here and is building on prior platform waves; most net new software revenue is coming from AI at both the application and infrastructure layers, and enterprise adoption is now translating into real usage and revenue, with app companies growing from zero to $100M revenue in a year or two.
Existing software categories going AI-native.
AI software can do a large share of human labor tasks, expanding the addressable market from software into the much larger labor market; the value-to-cost equation enables businesses to deploy AI where humans were too costly, but defensibility still requires becoming a system of record or vertical operating system.
Rillet AI-native ERP for inflection points.
Rillet is an AI-native NetSuite alternative that closes books and has 50 AI features built in, benefiting when companies hit inflection points requiring multi-entity or multi-currency ERP and are choosing new systems.
Incumbent software monetizes AI with hostages.
Incumbents with systems of record and hostage customers can monetize AI by charging for new features or outcomes, making their businesses stronger; examples include Workday charging for reference checks, Intuit/QuickBooks charging per collection, and SAP, Adobe, and Bill.com benefiting from AI.
Eve owns plaintiff legal AI workflow.
Eve is an AI-native plaintiff legal platform that owns the end-to-end workflow from intake to outcome; contingency-fee plaintiff attorneys align with AI productivity, and Eve's proprietary outcome data creates a compounding data moat and expands the market by making smaller cases economical.
Salient AI collections lift revenue 50%.
Salient uses AI to handle auto loan servicing and collections, collecting 50% more while lowering cost and ensuring compliance across evolving state and federal statutes; it becomes a system of record and data moat from millions of calls.
Vertical software plus financial services wins.
Vertical software companies can become very large by embedding deeply in an industry and expanding into financial services, making them hard to displace; Toast exemplifies this by adding lending and payment processing to restaurant software.
Toast's vertical OS plus payments compounds.
Toast was underestimated as a restaurant software company but became a valuable public company by adding financial services like lending and payment processing, creating a sticky vertical operating system that traditional payment processors cannot easily append software to.
FlightAware owns aggregated flight data.
FlightAware aggregates freely available ADSB transponder data through a global antenna network and owns a unique historical and real-time dataset that general AI models cannot answer; AI increases the value of this data.
PitchBook data enables AI finished product.
PitchBook owns historical private funding-round data, such as Series B prices since 1992, that general models lack; with AI it can deliver finished memos and analysis, supporting higher pricing than raw subscriptions.
LexisNexis data plus AI delivers answers.
LexisNexis owns proprietary legal and financial data and can use AI to deliver finished fraud or transaction decisions rather than raw data, making it a bullish incumbent if it is the only or best source.
DomainTools owns historical WHOIS data.
DomainTools has historical WHOIS domain ownership data going back to 1998 that is not available elsewhere; AI makes this proprietary data much more valuable for finished use cases.
Open Evidence owns medical journal data.
Open Evidence has exclusive licensing to the New England Journal of Medicine and other medical journals, creating a proprietary data walled garden; with a ChatGPT-like interface it delivers evidence-based answers that general models cannot, making it valuable to doctors who use it weekly.
vLex legal data plus AI quintupled revenue.
vLex digitized every legal record in Spain and other legal data, then added AI to deliver finished legal memos incorporating Spanish case law, quintupling revenue and becoming the only solution for certain cross-border legal work.
Ask Leo owns proprietary contract data.
Ask Leo is a procurement product that owns a proprietary trove of old contracts, such as Deloitte contracts, that general models cannot access, enabling better negotiation and pushback for enterprise procurement.
MSP AI roll-up scales digital client intake.
In IT managed services, much work is remote and digital and the market is $100B; buying an MSP and using AI to ingest more clients can scale better than buying hundreds of fragmented local services firms.
Consumer AI follows same app playbook.
The same AI application patterns are playing out in consumer: AI-native replacements for traditional categories like Krea versus Photoshop, new categories like Eleven Labs in voice, and proprietary data plays like Slingshot; the firm applies a consistent approach across these consumer AI themes.
Krea is AI-native Photoshop for newcomers.
Krea is the AI-native Photoshop, a strong product with AI primitives built in, and it is being chosen by young designers adopting their first design tool, displacing traditional Photoshop for new users.
Eleven Labs created voice AI category.
Eleven Labs created the voice and audio model category that barely existed five years ago, provides both consumer and enterprise SKUs, and vertically integrates to go after a large opportunity quickly.
Slingshot proprietary therapy data powers Ash.
Slingshot uses an AI scribe for existing therapists to collect proprietary therapy data, trains a foundation model, and sells a consumer AI therapist named Ash directly; this data advantage differentiates it from general models and supports high pricing.
Salesforce vulnerable to AI-native CRM.
Salesforce is vulnerable because users dislike it and a new AI-native greenfield CRM can outsell it; this is a specific example of how AI-native startups can attack incumbent software in categories where customers are unhappy.
This a16z video, published January 19, 2026,
features Alex Rampell, David Faber, Anish Acharya
discussing AI Application Software, AI-SECTOR, Rillet, WDAY, SAP, ADBE, BILL, INTU, Eve, Salient, Vertical Software, TOST, FlightAware, PitchBook, RELX, DomainTools, Open Evidence, vLex, Ask Leo, Managed service provider (MSP) AI roll-up, Krea, Eleven Labs, Slingshot, CRM.
21 trade ideas extracted by AI with direction and confidence scoring.
Speakers:
Alex Rampell,
David Faber,
Anish Acharya
· Tickers:
AI Application Software,
AI-SECTOR,
Rillet,
WDAY,
SAP,
ADBE,
BILL,
INTU,
Eve,
Salient,
Vertical Software,
TOST,
FlightAware,
PitchBook,
RELX,
DomainTools,
Open Evidence,
vLex,
Ask Leo,
Managed service provider (MSP) AI roll-up,
Krea,
Eleven Labs,
Slingshot,
CRM
