SaaS Multiples Reset🚨, How to Get Better Outputs from ChatGPT💡, What Is Your Startup Actually Worth?🎯
- 01🔴 SaaS Valuations Are Structurally, Not Cyclically, Repricing
- 02🤖 AI Infrastructure Is Abstracting Away Developer Complexity
- 03🧠 Enterprise AI Is Creating Compounding Learning Loops
- 04💸 Venture Funding Is Concentrating, Not Growing
- 05🌽 Geographic Arbitrage Opportunity in the Midwest
SaaS Multiples Reset, ChatGPT Outputs, Startup Valuation
1. Key Themes
🔴 SaaS Valuations Are Structurally, Not Cyclically, Repricing
The pressure on SaaS multiples is not a temporary market correction — it reflects a fundamental shift in how software value is priced and consumed.
"Software companies are trading below broader market benchmarks as growth expectations get repriced. AI-native pricing models are pulling budget away from seat-based subscriptions, challenging core assumptions."
🤖 AI Infrastructure Is Abstracting Away Developer Complexity
The operational burden of building AI-powered products is collapsing. This lowers barriers to entry and accelerates time-to-ship for agentic workflows.
"Teams no longer need to build sandboxes, memory, and recovery systems before shipping usable agent workflows. Managed Agents package the operational layer so developers focus directly on behavior and execution."
🧠 Enterprise AI Is Creating Compounding Learning Loops
The next moat in enterprise software won't be features — it will be proprietary behavioral training data generated by everyday employee decisions.
"Enterprise systems tracked outcomes but ignored the reasoning behind them, leaving no learning layer to compound over time. AI agents now log edits and overrides as structured traces, turning everyday decisions into a reusable training system."
💸 Venture Funding Is Concentrating, Not Growing
Record headline numbers mask a hollowing out of broad-based deal activity — capital is pooling at the very top.
"Global funding surged to record levels but was heavily concentrated in a few mega rounds led by OpenAI. Deal activity and investor participation declined, signaling a shift toward fewer firms writing larger checks."
🌽 Geographic Arbitrage Opportunity in the Midwest
A data-backed case is emerging for the Midwest as an undervalued startup ecosystem — high founder activity, thin venture coverage.
"A data driven report highlights the Midwest's growing startup density and rising founder activity across key cities and sectors. Despite momentum, a persistent venture capital gap continues to limit how far these companies can scale."
2. Contrarian Perspectives
Anthropic May Be the More Disciplined AI Bet — Not OpenAI
While OpenAI dominates headline funding rounds, Anthropic is quietly building a more capital-efficient path to profitability by combining enterprise revenue growth with lower compute burn.
"The company reached $30B ARR quickly by leaning into enterprise demand while keeping compute costs disciplined. Lower projected training spend positions it to reach profitability earlier despite competing at similar scale." The contrast with OpenAI — the anchor of the "record" VC funding concentration — is stark. Anthropic appears to be optimizing for unit economics while OpenAI optimizes for scale at cost.
Online Presence Now Precedes — and Determines — In-Person Access
Conventional startup wisdom prizes warm introductions and in-person relationship building. The article (citing a16z) argues the causality has reversed: your digital footprint now grants you the in-person meeting, not the other way around.
"Social graphs built on platforms and group chats now determine access more than traditional introductions. In-person events function as extensions of ongoing online context rather than initial connection points."
The Most Dangerous AI Capability Is Being Deliberately Kept Scarce
Rather than racing to deploy its most advanced model, Anthropic is restricting access to a system capable of autonomously discovering and exploiting security vulnerabilities — and using it defensively instead.
"Anthropic is limiting access to its most advanced system due to its ability to discover and exploit vulnerabilities autonomously. Through Project Glasswing, it is being deployed to secure infrastructure before threats can emerge." This is a notable departure from the "ship fast" AI norm and signals that capability ceilings are arriving faster than public discourse acknowledges.
3. Companies Identified
| Company | Description | Why Mentioned | Quote |
|---|---|---|---|
| Anthropic | AI safety-focused foundation model company | Multiple mentions: operational efficiency, enterprise ARR, and restricted advanced model deployment | "Reached $30B ARR quickly by leaning into enterprise demand while keeping compute costs disciplined." |
| OpenAI | Leading AI lab and consumer/enterprise AI platform | Anchor of record VC funding concentration; leaked cap table circulating | "Global funding surged to record levels but was heavily concentrated in a few mega rounds led by OpenAI." |
| Polymarket | Prediction market platform | Used to surface low market confidence in near-term Mythos AI release | "Traders currently assign a low probability to a near-term release based on available information." |
| Vanta | AI governance and compliance automation | Newsletter sponsor; highlighted for AI governance framework | "6 clear steps to align with recognized standards, define ownership, and integrate AI risk into your existing compliance program." |
| Attio | Modern CRM for startups and VCs | Endorsed by newsletter author as a tool used personally | "The CRM used by both startups and VCs (including me)." |
| SiFive | RISC-V semiconductor company | Largest deal featured: $400M Series G | "Secured $400M in Series G funding to accelerate RISC-V semiconductor innovation." |
| Eclipse | Deep tech VC firm | Raised $1.31B for manufacturing, robotics, and energy | "Raised $1.31B for two funds focused on manufacturing, robotics, and energy (~$10B AUM total)." |
| Collide Capital | Fintech/supply chain/future-of-work VC | Closed $95M Fund II | "Closed a $95M Fund II to invest in fintech, supply chain, and future-of-work startups." |
| Chapter | Medicare advisory platform | Raised $100M Series E — notable health-tech deal | "Raised $100M in Series E funding to expand its Medicare advisory platform." |
4. People Identified
| Person | Description | Why Mentioned | Quote |
|---|---|---|---|
| Ruben Dominguez | Author of The VC Corner newsletter; investor and entrepreneur | Curator and primary voice of the newsletter | "For sponsorship opportunities across this newsletter and LinkedIn (290k followers), email: ruben@thevccorner.com" |
| Jaya Gupta | Contributor/analyst cited in the enterprise AI feedback loop piece | Credited with the insight on AI agents turning decisions into training data | "AI agents now log edits and overrides as structured traces, turning everyday decisions into a reusable training system." [Jaya Gupta] |
5. Operating Insights
Prompt Structure — Not Model Quality — Is the Real Bottleneck for AI Output
Most teams are leaving significant productivity on the table because they blame the model when the issue is the input. Role assignment and reusable context blocks are low-effort, high-return fixes.
"Most poor results come from vague inputs rather than model limits, making prompt structure the real bottleneck. Simple tactics like role assignment and reusable context blocks shift output from generic to tailored."
Founders Must Build Valuation From Multiple Methods, Not Gut Feel
Entering a fundraise with a single number — especially one anchored to emotion or peer comparisons — immediately undermines negotiating credibility when investors probe assumptions.
"Founders often anchor on arbitrary numbers, weakening their position the moment scrutiny begins. Using multiple valuation methods creates a defensible range and reframes discussions around assumptions."
Build Your Online Presence Before You Need the Room
For founders seeking to access top-tier investors and operators in dense ecosystems, the article signals that digital relationship-building is now the prerequisite — not the follow-up — to in-person access.
"Social graphs built on platforms and group chats now determine access more than traditional introductions. In-person events function as extensions of ongoing online context rather than initial connection points."
6. Overlooked Insights
Prediction Markets as a Due Diligence Tool for AI Product Timelines
The mention of Polymarket pricing in low probability for a near-term Mythos AI release is a brief but notable signal: incentive-aligned prediction markets may be a more reliable forecasting mechanism for AI product timelines than press releases or founder roadmaps.
"Market pricing reflects expectations of delays from participants with incentives tied to accuracy." For investors evaluating AI companies, tracking relevant prediction markets could serve as a real-time sentiment and timing signal that traditional research misses.
AI Governance Is Becoming a Fundraising and Compliance Prerequisite
Tucked inside a sponsored section, the framing around Vanta's AI Governance Checklist points to a quietly emerging requirement: as AI adoption accelerates, governance documentation is transitioning from a nice-to-have to a threshold criterion for enterprise sales and investor due diligence.
"AI adoption is accelerating fast, but so are governance expectations." Startups that embed AI risk management early will have a structural advantage in regulated-industry sales cycles and future compliance audits.