Anthropic and OpenAI Set Up AI Services Firms as CIOs Cut Their Vendor Lists
Updated: 6 days ago
Title: Anthropic and OpenAI Set Up AI Services Firms as CIOs Cut Their Vendor Lists
Date: 9 May 2026
Type: Blog
Author: SAASiQ (contact@saasiq.ai)
Word count: 1299 words
Reading time: 5 min
Published: 09-05-2026
Anthropic announced on 4 May that it was setting up an enterprise AI services firm with Blackstone, Hellman & Friedman and Goldman Sachs, and Bloomberg reported the same day that OpenAI had finalised a similar venture with a group of private equity firms. Reuters reported on 5 May that OpenAI's venture was in advanced talks on three acquisitions. Buyers, meanwhile, are trimming their supplier lists: in a Redpoint Ventures survey of 141 CIOs in March, 54 per cent said they were actively consolidating vendors and 3 per cent expected AI to leave them with more.
What Anthropic and OpenAI set up
Anthropic's new firm is valued at $1.5 billion, with $300 million committed by each of Anthropic, Blackstone and Hellman & Friedman, and General Atlantic, Leonard Green, Apollo, GIC and Sequoia Capital are also backing it. It is aimed at mid-sized companies, and Anthropic names community banks, manufacturers and regional health systems as examples. Anthropic's applied AI engineers will work alongside the firm's own engineers to find where Claude fits, build the tools and support them afterwards, and the firm joins the Claude Partner Network next to Accenture, Deloitte and PwC. Krishna Rao, Anthropic's chief financial officer, said enterprise demand for Claude "is significantly outpacing any single delivery model".
OpenAI's version, called The Deployment Company, has raised more than $4 billion from 19 investors including TPG, Brookfield Asset Management, Advent and Bain Capital, according to Bloomberg. The deal values it at $10 billion before the new money, and OpenAI keeps majority ownership and control.
Reuters reported that most of the capital in both ventures is expected to go on buying engineering services and consulting firms, and that each plans to take on hundreds of engineers and consultants. Jon Gray, Blackstone's president, said the aim was to ease one of the main constraints on enterprise AI adoption, which is a shortage of skilled implementation partners. OpenAI was already buying at a steady rate: Crunchbase counted six acquisitions by 25 March, against eight in the whole of 2025.
What CIOs say about vendor numbers
Redpoint's survey, run in March, put numbers on the buyer's side. Alongside the 54 per cent consolidating and the 3 per cent expecting more vendors, 45 per cent of CIOs said their AI budgets were coming out of existing software line items rather than new money.
Where the money comes from an existing line, a new AI product has to displace something already on the books, and the incumbent supplier that adds AI features to its own product is competing for the same budget as the AI-native newcomer.
Gartner estimated in June 2025 that only about 130 of the thousands of vendors claiming agentic AI offer real agentic features, a practice it calls 'agent washing', and predicted that over 40 per cent of agentic AI projects will be cancelled by the end of 2027 because of escalating costs, unclear business value or inadequate risk controls.
Software companies buying AI controls
The large software vendors are consolidating too, by acquisition. ServiceNow completed its $7.75 billion cash purchase of Armis on 20 April, a month after closing its acquisition of Veza. Armis tracks nearly 7 billion connected devices across IT, operational technology, medical equipment and cloud, and Veza maps the permissions held by every human, machine and AI agent identity across a customer's systems. ServiceNow says the two together more than triple its addressable market for security and risk products.
On 30 April Palo Alto Networks agreed to buy Portkey, an AI gateway that gives companies one place to manage and protect their AI agents and, in Palo Alto's words, processes trillions of tokens a month. It will become part of Prisma AIRS, Palo Alto's security platform for AI. The price was not disclosed and the deal is expected to close in Palo Alto's fiscal fourth quarter. Earlier, on 17 March, IBM completed its $11 billion purchase of Confluent, the data streaming platform used by more than 6,500 enterprises, which IBM is selling as the source of real-time data for AI models and agents.
Microsoft made Agent 365 generally available on 1 May at $15 per user per month, and it is included in Microsoft 365 E7. Microsoft describes it as a control plane for observing, governing and securing agents, with a registry that can sync, in preview, with agents on AWS Bedrock and Google Cloud.
Where the money is going
Gartner's forecast of 22 April put worldwide IT spending at $6.31 trillion in 2026, up 13.5 per cent, and data centre systems at $788 billion, up 55.8 per cent. John-David Lovelock, the Gartner analyst behind the forecast, said AI infrastructure and generative AI software had been revised up substantially, while devices were held back by cost and price pressures.
In their first-quarter results at the end of April, Alphabet, Amazon, Microsoft and Meta set out plans to spend a combined $725 billion on capital projects in 2026, up 77 per cent from $410 billion in 2025, according to a Financial Times tally of their guidance.
Pilots and production
Deloitte's State of AI in the Enterprise report, published on 21 January and based on a survey of 3,235 business and IT leaders in 24 countries in August and September 2025, found that 25 per cent of organisations had moved 40 per cent or more of their AI pilots into production. Another 54 per cent expected to reach that level within three to six months.
McKinsey's State of AI survey, published in November 2025 from 1,993 responses, found 62 per cent of organisations at least experimenting with AI agents and 23 per cent scaling an agentic system somewhere in the business. Gartner predicted in August 2025 that 40 per cent of enterprise applications will include task-specific AI agents by the end of 2026, up from less than 5 per cent in 2025. Many of those agents will arrive inside applications a company already licenses.
Deloitte's respondents named insufficient worker skills as the biggest barrier to integrating AI into existing workflows. Fewer than half were making significant changes to their talent strategies, and the most common response, from 53 per cent, was educating the wider workforce to raise AI fluency. Deloitte also found that 84 per cent had not redesigned jobs or the nature of work around AI.
Governance for agents
Nearly three in four companies in Deloitte's survey, 74 per cent, plan to deploy agentic AI within two years, but only 21 per cent report a mature model for governing autonomous agents. Deloitte's advice is to start with lower-risk uses, set clear limits on what an agent may decide without a person's approval, and keep audit trails that capture the full chain of an agent's actions. It also found that companies whose senior leadership actively shapes AI governance get significantly greater business value than those that leave the work to technical teams alone.
Oracle customers get agents through their existing licence. On its third-quarter earnings call on 10 March, Oracle said it had delivered well over 1,000 AI agents inside its applications, at no additional cost. The 26B quarterly update reached test environments on 1 May, and its readiness notes show the new agentic applications stay off until an administrator switches them on. An agent in Fusion works within the access its user already has, so the role design done for an implementation is also the security design for its agents.
SAASiQ's view is that Fusion customers should review the agents already in their licence, and who has access to them, before adding another AI supplier, since Redpoint's figures suggest the money for a new one would come out of an existing software line.
Oracle's 26B update is due in production on 15 May, and neither Anthropic nor OpenAI has yet named the services firms their ventures intend to buy.
SAASiQ - Intelligent Solutions for SaaS ©
