AI Labs Ditch Models for Consultants as SaaS Market Loses $Billions in Value

TLDR:

• OpenAI raised $4B to launch a consulting firm competing directly with Deloitte, PwC, and Ernst & Young
• Anthropic countered with a $1.5B consulting venture backed by Blackstone and Goldman Sachs within weeks
• Goldman Sachs reports software P/E multiples fell to 20x, the lowest absolute reading recorded since 2014
• 88% of organizations running AI agents reported at least one security incident over the past twelve months

The enterprise software sector is undergoing a structural shift as artificial intelligence reshapes how businesses buy and deploy technology.

Valuations across public SaaS companies have declined sharply, with 90% of stocks trading 30–80% below their 52-week highs.

Meanwhile, leading AI labs are moving beyond model development into hands-on consulting. The combined moves signal a broader realignment of where value is created in the software industry.

Low-End SaaS Tools Are Being Replaced by AI Workflows

The first segment to feel the pressure has been lightweight, single-function software products. Tools priced around $49 per seat per month are losing ground to AI agents that replicate their functions automatically.

Businesses no longer purchase a dedicated tool for a narrow task — they describe what they need, and AI builds and executes it.

Milk Road AI noted on X that “the low end of the market is basically finished,” citing investor Chamath Palihapitiya’s recent diagnosis of the sector.

The seat-based pricing model that built the SaaS industry does not translate to this new transaction type. As a result, the economic foundation of many small business software products is eroding.

Goldman Sachs data reflects the broader damage. Software forward price-to-earnings multiples dropped from 35x to 20x — the lowest level since 2014.

That multiple is also the smallest premium to the S&P 500 since 2010, pointing to a sector-wide reassessment of growth prospects.

Mid-market point solutions without proprietary data assets face a similar trajectory. Products that lack defensible data flywheels or deep vertical integration are increasingly vulnerable to displacement by general-purpose AI systems.

AI Labs Are Now Competing Directly With Top Consulting Firms

At the high end of the market, the challenge is different — but no less serious. OpenAI recently raised $4 billion from investors including TPG, Brookfield, Bain, and McKinsey to launch a consulting division. The venture targets direct competition with firms like Deloitte, PwC, Ernst & Young, and Accenture.

The structure of the deal is notable. Investors were guaranteed a 17.5% annual return — roughly $700 million per year — from a company projected to lose $14 billion in 2026.

The move came after OpenAI’s enterprise LLM market share dropped from 50% to 25% between late 2023 and mid-2025, with Anthropic rising to 32%.

Anthropic responded almost immediately with a competing $1.5 billion consulting venture, backed by Blackstone, Goldman Sachs, and Hellman & Friedman.

Together, the two labs committed $5.5 billion to human-powered enterprise deployment within a single month. The scale of that spending reflects how difficult real-world AI implementation remains.

Data supports that difficulty. Some 88% of organizations running AI agents reported a security incident in the past year.

Additionally, 42% of C-suite executives said AI adoption is generating internal organizational conflict. Average year-one implementation costs through consulting run $228,000 — nearly three times the platform-based alternative.