Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
About the Author
About the Author
About the Author

AI for AI: Innovate on Business Models – Destroy to Create Funding, Share Risk and Capture Value: What is State of Art and What Next?

Being Human for a Better Tomorrow in the Age of AI
  • March 7, 2025
  • 38th edition
AI for AI Innovate on Business Models Destroy to Create Funding Share Risk and Capture Value What is State of Art and What

Where’s the Money for AI?

The Economic Squeeze and Runaway AI Costs

We face a volatile, inflationary global economy and potential cost runaways

1. Volatile/Unpredictable Performance and Budgets:

  • Q2 FY24 saw India’s GDP growth forecast at 6.4% but delivered only 5.4%

  • 20% of S&P 500 companies and 57% of NIFTY companies missed earnings estimates in Q3 2024

2. Technology Inflation:

  • SaaS inflation jumped to 12.3% in 2024 (4x U.S. economic inflation)

  • Software vendors are implementing 30% annual increases as they embed AI into solutions

  • CIOs anticipate 25-35% rises in SaaS costs by 2025 as AI becomes ubiquitous

3. AI Runaway Costs:

AI projects consistently overrun budgets, with Gartner’s 1Q FY25 CIO report warning that costs can balloon 5-10x initial estimates, potentially consuming 35% of entire annual budgets.

The key reasons include:

  1. Scope creep and overambitious goals – A simple copilot balloons into a full workflow overhaul and legacy integration.

  2. Data preparation challenges – Cleaning and structuring data consumes 60-80% of project time/budget.

  3. Model and scaling complexities – Iterative fine-tuning and moving from pilot to production escalates costs.

  4. Infrastructure requirements – GenAI’s massive compute power demands strain budgets.

  5. Talent premiums – AI specialists command 30-50% above-market salaries.

Enterprises today face a three-body problem: legacy costs, AI chaos, and stagnant budgets tugging them into uncharted territory.

Survival demands obsession, new thinking, and agile navigation.

New Models for Funding AI and Capturing Value

Given these challenges, forward-thinking organizations need innovative approaches to:
  • Find new money – Review current tech stack and services to eliminate inefficiencies
  • Establish predictable pricing – Create clarity through usage caps or outcome-based models
  • Share risk with vendors – Ensure providers co-own performance targets
  • Maintain visibility – Implement clear cost metrics and real-time dashboards
  • Enable phased adoption – Scale or pause based on early ROI signals

Destroy to Create: Finding New Money for AI

Consider three strategic approaches to create new funding streams:

1. Audit & Slash

Conduct comprehensive tech stack audits with vendor support to identify redundant licenses or underused systems. Incentivize legacy vendors to help uncover savings.

2. Kill Legacy Business

Strategically sunset low-value offerings to reallocate resources toward high-impact AI initiatives. This approach transforms existing products or services, similar to Autodesk’s reimagining of CAD through AI.

3. Spinoff AI Labs

Establish internally funded or VC-co-invested AI labs to innovate and disrupt existing products with AI. This allows for agile experimentation without disrupting core operations, as demonstrated by HubSpot’s AI Lab spinoff and Baidu Research AI unit spinoff with VC co-investment (e.g., Apollo Go with $1.5B from VCs).

Smart Money: Managing Risks and Capturing Value

Once you’ve found the funding, invest it strategically with these approaches:
  • Hybrid Pricing Models – Implement structures combining fixed base fees with variable performance fees tied to business outcomes. This ensures predictable costs while aligning vendor incentives with your success.
  • AI Cost Control Dashboards – Deploy real-time monitoring of compute costs, data usage, and outcome progress. Set alerts for budget thresholds and benchmark against industry peers.
  • Phased Adoption with Exit Ramps – Start with low-cost pilots to validate ROI before scaling. Maintain exit options if targets aren't met, preventing runaway investments in underperforming initiatives.
  • Value-Based Partnerships – Structure agreements where vendors' compensation directly correlates with the value delivered, whether through revenue increases or cost savings.
  • Outcomes-Driven Contracting – Focus on business results rather than technology deliverables, with compensation tied to achieving specific, measurable goals that matter to stakeholders.

The most successful AI innovators aren’t just reimagining technology—they’re reinventing how value is created, shared, and captured.

As the sidebar illustrates, the disruptors featured in previous editions have all pioneered innovative business models alongside their technological breakthroughs.

Takeaways

AI-led Accelerated Innovation (AI for AI) continues to transform industries. AI technology and infrastructure are rapidly evolving, supporting super agency to deliver revenue and strategic value.

But it’s not only about technology. New business models are also emerging in the Age of AI. My hypothesis is that those who innovate on both technology and business models will become runaway leaders, leaving others far behind.

Being efficient is one path; being disruptive is another. Both deserve pursuit.

To drive disruption, ask these key questions:

  • What hard problems can you solve in your business and function? Ask Peter Thiel's question: What important truth do very few people agree with you on?
  • What value can be created by solving this problem?
  • How will AI enable the solution?
  • What internal and external partners are needed to plan and execute?
  • How can you be part of this change in your industry with your know-how and skills?
And now, two additional questions:
  • What can you do to maximize new money for AI initiatives?
  • What innovations in business models will optimize your AI investments?

Final Word

Across our five editions on AI for AI, we’ve seen how a small team at DeepSeek disrupted AI itself, and explored 37 AI and Tech-led innovators transforming FS, Pharma, Healthcare, Enterprise Software, and Physical AI.
More exciting examples await in upcoming editions.

In democratized AI, anyone, anywhere can achieve remarkable results with courage, intelligence, and tenacity.
And the time to act is now.

What a promising start to 2025—the first year of the next quarter-century.
Explore, join, and stay tuned for more!

Love to hear your comments, thoughts, and ideas.

Note:

The five aids to create an intelligent Promethean enterprise in the Age of AI are:
  • Supporting culture
  • Employee First Use Case Approach for Gen AI Implementation and Customer First for More Established Machine Learning/Data Sciences
  • Bi-Modal Exploration (AI Enhanced and AI Driven)
  • Save IT Opex, Invest First, Build evidence-driven ROI retroactively for Future
  • Do No Harm

Sidebar: Innovative Revenue Models from AI Disruptors

The innovators from editions 33–37 don’t just push tech—they rewrite how value is funded, shared, and captured. Here’s how they align across five bold models:

1. Risk-Sharing & Outcome-Based Models

Tie revenue to results—vendors and customers co-own the upside and downside.

  • Biofourmis (35th): Healthcare performance contracts cut readmissions 38%.
  • Zipline (35th): Public health metrics slash maternal mortality 40%.
  • Insilico Medicine (34th): Pharma milestone payments tied to AI-driven drug progress.
  • Kiva (33rd): Nonprofit AI lending scales via optional fees.
  • Exscientia (34th): Outcome-based pharma AI earns on clinical wins.
  • Covariant (37th): SaaS + per-pick fees for robotic AI, shared learning boosts value.
  • Upstart (33rd): AI lending ties revenue to 30% higher loan approvals.
  • BlackRock LifePath (33rd): AI-driven dynamic investments for guaranteed benefits.

2. Hardware-Embedded AI Value

Bake AI into the gear—no extra fees, just smarter products.

  • Apple Intelligence (37th): Enhances iPhone value with seamless AI.
  • Tesla FSD (37th): Full Self-Driving embeds autonomy in cars.
  • Hailo-8 (37th): Edge AI chips power efficiency without standalone costs.
  • Chef Robotics (37th): Food prep robots fuse hardware and AI smarts.
  • Sanctuary AI (37th): Humanoid robots deliver value through physical AI.
  • Noteworthy AI (37th): Grid inspection tech embeds AI in hardware.
  • NVIDIA Jetson (37th): Developer kits bundle AI power for edge innovation.

3. Open Ecosystem Monetization

Free hooks draw crowds; premium tiers cash in.

  • DeepSeek (33rd): Open-source AI models with enterprise upsells.
  • LangChain (36th): Community-driven AI tools, paid features for scale.
  • Weights & Biases (36th): MLOps platform blends free trials with pro tiers.
  • Hugging Face (36th): Open-source LLMs, enterprise solutions for profit.
  • Block’s Goose (33rd): Open API ecosystem drives adoption, monetizes scale.

4. Subscription Transformations

Shift from one-time buys to recurring AI-powered streams.

  • Autodesk (36th): CAD licenses morph into AI-enhanced SaaS.
  • Replit (36th): Tiered coding subs scale with AI features.
  • Jasper (36th): Content AI offers usage-based subscriptions.
  • HubSpot AI Labs (36th): Self-disrupts legacy revenue with AI subs.
  • Figma AI (36th): Design tools pivot to AI-driven recurring plans.
  • Runway (36th): Video creation shifts to subscription-based AI access.

5. Service as a Software

Flip the script—charge for outcomes, not just tools.

  • Betterment (33rd): Wealth management scales via low-cost, AI-driven personalization.
  • Woebot (35th): Affordable AI therapy subs deliver mental health access.
  • Moxi (37th): Robot-as-a-Service leases with continuous AI upgrades.
  • Robinhood (33rd): Trading as a service with AI-driven insights.
  • Wealthfront (33rd): Automated investing as a service, AI at the core.
  • Figure (37th): Humanoid robotics as a service for enterprise tasks.
“AI is a language. Treat it like one: practice, iterate, and mind your grammar prompts, assumptions, and verification.”

II. Questioning / Asking

Good conversations flow from well‑sequenced questions—topical, simple, coherent, cohesive.
LLM Conversation Example 1
Q: What are empirical judgments?

A: Empirical judgments are based on observation, experience, or experimentation.
Q: What are moral judgments?

A: Moral judgments are based on ethical principles and values.

IV. Judgment

“The structure of a language affects its speakers’ worldview and cognition.”
— Henry Hazlitt
“The art of questioning is the source of all knowledge.”
— Thomas Berger
Share: