OpenAI Is Going Public. Here Is What That Means for Your Business.
OpenAI's trillion-dollar IPO filing, a study that reframes the AI adoption debate, and a pricing war that has a narrow window.
This week, three things happened in AI that are easy to miss in isolation and important to understand together.
The biggest AI company in the world filed paperwork to go public. A research firm published data showing that most large companies got AI wrong. And three competing platforms cut prices simultaneously in ways that will not last.
Each story on its own is interesting. Together, they tell you something about where the industry actually is and what that means for how you run your business.
1. OpenAI Is Filing for an IPO, and the Numbers Are Revealing
On May 22, OpenAI filed confidential IPO paperwork with the SEC, targeting a public listing as early as September at a valuation of up to one trillion dollars. Goldman Sachs and Morgan Stanley are leading the deal.
The business headlines focused on the valuation. The more useful thing to understand is what the financials reportedly look like underneath it.
OpenAI is generating roughly two billion dollars per month in revenue and has an annualized run rate of approximately 25 billion dollars. It counts around 50 million consumer subscribers and nine million business users. By any measure, those are extraordinary numbers for a company that launched its first commercial product in late 2022.
But the company is also reportedly losing more than a dollar for every dollar it earns, with no expectation of positive cash flow until 2030. The infrastructure costs -- the compute required to run models at this scale -- are that large.
Why does this matter for a service business owner who is not buying OpenAI stock?
Because going public fundamentally changes how a company makes decisions. Right now, OpenAI is accountable to private investors and a nonprofit board. Once it is publicly traded, it answers to shareholders every quarter. That creates pressure to grow revenue predictably, trim costs, and prioritize enterprise contracts over consumer experimentation.
In practical terms: the tools you use today are about to be priced and prioritized by a public company with fiduciary obligations. That is not necessarily bad -- public companies often invest heavily in enterprise reliability and support. But it is a different relationship than the current one, and it is worth factoring into any multi-year decisions you are making about which AI platforms your business depends on.
For additional context: Anthropic is reportedly targeting its own IPO in October at a valuation above 900 billion dollars. Two of the three dominant AI platforms going public in the same year is without precedent, and the financial transparency it forces will be clarifying for everyone.
2. The Gartner Study That Buried the Real Finding
A Gartner survey of 350 global business executives released this month found that 80 percent of companies that piloted AI or automation technology reduced their workforce as a result.
That number circulated widely and generated a lot of concern. But the finding that got much less attention is the more important one: those workforce reductions had no measurable correlation to higher return on investment. Companies that cut jobs because of AI did not see better financial performance. The cuts happened. The returns did not follow.
What Gartner found, looking more closely, is that the companies struggling were the ones that moved directly to headcount changes without redesigning the workflows underneath. They automated the output without rethinking the process. They reduced the team without rebuilding how work gets done. And so they ended up with fewer people running the same broken systems, which produces the same results at lower cost but not materially better ones.
This is not an abstract finding. It maps to what I see regularly in implementation work. Businesses that treat AI as a cost-cutting tool, rather than a process-rebuilding tool, typically see modest efficiency gains that plateau quickly. Businesses that use AI as a reason to rethink how work actually flows -- which tasks require human judgment, which do not, where the bottlenecks actually live -- tend to see compounding returns over time.
The Gartner study is not an argument against AI adoption. It is an argument that adoption without implementation planning produces the same problems as not moving at all. The tool does not do the work. The redesign does.
3. The AI Pricing War and What to Do With It
At Google's I/O conference on May 19, the company announced it was cutting its top-tier AI Ultra subscription from 250 dollars to 200 dollars per month, and introducing a new entry-level AI Ultra tier at 100 dollars per month.
The same week, OpenAI offered enterprise users two free months of access to its Codex coding tool if they switched within 30 days. Anthropic increased usage limits for Claude by 50 percent for paid users through July 13.
Three major platforms making simultaneous pricing moves is not a coincidence. It is competition. And for buyers, genuine competition is good news in the short term.
The honest read, though, is that pricing wars in this industry tend to compress before they widen. What you are seeing is companies fighting for paying customers, not necessarily signaling that AI tools are becoming permanently cheaper. The free months and the bumped limits are retention and acquisition plays. The underlying economics -- which include the same infrastructure costs that are keeping OpenAI unprofitable -- have not changed.
What this week's moves do signal is that there are more options at more price points than there were six months ago. If you have been hesitant to upgrade a tool because the pricing felt steep, this is a reasonable moment to revisit that. If you have been letting usage limits constrain what you build, this is a window to move faster on something you have been planning.
One additional note from I/O worth flagging: Google announced that its Search redesign -- the biggest change to the search interface in 25 years -- now includes agents that can call businesses on behalf of users for categories like home repair, beauty, and pet care. Earlier this year, research tracking 50 keywords found that pages in the top three search results saw click-through rates drop between 18 and 34 percent once AI-generated answers appeared above them. If your business relies on search traffic to bring in clients, this is a slow-moving but consequential story to watch.
What This Means
The pattern across all three stories is the same: AI is growing up.
An IPO forces financial transparency that venture funding never required. A study generating accountability data about what works and what does not is how any market matures. A pricing war with three well-funded competitors is what happens when a category stops being dominated by novelty and starts being evaluated by actual value delivered.
For the businesses that moved thoughtfully -- that treated AI like infrastructure and rebuilt workflows instead of just adding tools -- this moment is validating. For the businesses that experimented without a plan, the accountability that is arriving is going to be clarifying.
The window to get intentional is not closing. But the cost of not having a plan is becoming more visible.
What Business Owners Should Actually Do
Audit your AI subscriptions before the pricing window closes. The current promotions from Google, OpenAI, and Anthropic are time-limited. If you have been meaning to upgrade or consolidate, the next 60 days are a reasonable time to act.
Before you cut anything, redesign first. If you are considering using AI to reduce labor costs, the Gartner finding is a useful check: map the workflow before you change the headcount. The redesign is the investment, not the reduction.
Get clear on which AI tools are core to your business versus which are experiments. OpenAI going public means the tools you depend on are entering a new pricing and prioritization environment. Know what you cannot afford to lose access to, and make sure those relationships are formalized.
Watch the Google Search shift. If search is part of how clients find you, the AI Mode changes announced at I/O this week are worth understanding before they affect your traffic.
Treat AI like infrastructure, not a subscription. The businesses struggling in the Gartner data were the ones that approached AI as a cost line to optimize rather than a capability to build. The framing matters more than the tool choice.
Chantal Emmanuel is co-founder of BAMPT, an AI automation implementation firm for service businesses, and CTO of LimeLoop and Gatheron.