Generative AI is entering its most competitive phase yet. On June 1, Anthropic filed confidential IPO documents with the SEC. A week later, OpenAI announced that it had done the same. Meanwhile, SpaceXfresh option to get AI Coding Launch Cursor with a reported valuation of $60 billion, it is now signaling enterprise ambitions.
Two companies, a week apart, a combined valuation north of $1.8 trillion, and yet a single question hangs over both filings that the prospectuses can’t quite answer: when each model is “smart enough,” what exactly are investors buying? As standards priorities narrow and core model capabilities converge, the gap is shifting from intelligence and distribution to something much harder to replicate: personality.
Branding in AI is behavior
In AI—perhaps more than any category where the product itself is abstract and unseen—branding is something stronger than a logo and color palette. It’s behavior. Two models can arrive at the same answer and feel completely different getting there. One may be short and transactional, the other patient and exploratory. This change, repeated daily across millions of interactions, becomes identity. And faith is what is at stake in that replay. You don’t trust a standard, you trust a pattern of behavior you know.
Anthropogenic it seems to treat that pattern as an engineering objective, not an accident. The company hires a “head of personality alignment,” philosopher Amanda Askelland in 2026 published Claude’s “constitution”—a roughly 20,000-word public document listing Claude’s values: safety, ethics, compliance with guidelines, and then helpfulness. It is against this background, the designed behavior of the brand, that the divergence between OpenAI and Anthropic gets interesting as both move toward IPO-era models and litigation enterprise engagements.
OpenAI’s bet is distribution on a planetary scale: its valuation rose from $86 billion to over 850 billion dollars in just over two years, and ChatGPT has become a verb, a habit for him billion people now use it every month to write, think and search.
Habits are contagious and being a default is worth more than being a little smarter. But this strategy carries a built-in tension: the more people rely on a product to draft emails, hide doubts, and shape decisions, the more its personality and commercial incentives come under scrutiny.
OpenAI has lived this out in public. A 2025 update was made GPT-4o significantly nicerconfirming users’ suspicions and, in one widely reported case, telling a user who had stopped taking the medication and was hearing rumors that they were “telling their truth”. OpenAI pulled the update Within days, acceptance of the change had undermined people’s faith in the product – believe it can’t outrun its way once it’s spent.
His fix has been to make the personality tunable rather than fixed. The GPT-5 ships with four selectable personality modes, and the default tone has since been tweaked again after complaints that it felt too formal. None of this is unusual for a high-growth tech company. But in AI, it faces another concern: scale attracts attention. You don’t automatically gain trust.
Instead, Anthropic has chosen a quieter path. It’s clearly a commercial player – it’s raised big sums and signed big deals – but it’s chosen to present Claude as a specialist tool for people who build things, rather than a mass-market consumer brand.
This positioning appears in the current usage. Claude has become part of enterprise and start-up engineering teams leading agent software development. I’ve seen senior engineers orchestrate fleets of Claude agents in parallel — each writing, testing, and running code — multiplying output without increasing headcount. This is a firm that rebuilds its operating model around a tool it trusts to behave consistently, driven by throughput. For enterprise buyers, this consistency is insurance risk. A model that behaves predictably is one that legal, compliance and engineering leadership can all sign off on.
Trust, it turns out, is a core brand value, predictability is a feature, and in the enterprise market – where erratic production can translate into real costs – consistency of behavior reads as professionalism. Unlike a benchmark, these are traits that cannot be replicated overnight, and we find that they are best achieved when product and brand development teams work together in close synergy.
How personality hardens in a ditch
Here the soft quality of personality turns into something closer to market structure. When a company builds its engineering pipeline, customer workflows, or internal knowledge about the specific behavior of a model, the shift stops being about price and starts being about realigning how the organization works. That’s the payoff of commoditization: once any model is smart enough, the only thing left to spark is trust, and trust is contagious in a way that raw skill never was.
You can migrate data or renegotiate a contract; you cannot easily replace the accumulated adaptation between a team and a tool that has learned its rhythms and that they have learned to predict. The gap is not the pattern. It is the relationship that the model has earned. Faith, embedded in daily work, becomes infrastructure, and infrastructure is precisely what no one tears up over a two-point difference.
None of this unfolds in a vacuum. The Pope has warned that AI has no conscience, a reminder that the public is questioning not just whether these tools are capable, but whether they can be trusted with anything that matters. That concern quickly moves from pulpits and op-eds to boardrooms and compliance inboxes—the same deficit both companies are now racing to close. Whoever closes it with enterprise buyers, not just headlines, wins the most consistent prize.
The other lever
IPO filings will be divided on revenue multiples and growth curves, but the most interesting story lives off the charts. When intelligence is everywhere, the differentiator is personality, the consistent, reliable, habit-forming behavior that makes a person, or an entire engineering organization, build their workflow around one system instead of another.
That’s the crucial question now: whose intelligence is felt in compliance with regulation, reputational risk, and the day-to-day realities of running a business.
OpenAI and Anthropic have offered completely different answers. As listings approach and contracts are signed, we’re ready to find out which one companies and investors believe. In an age where intelligence is free, trust is the only thing left to pay a premium for, and enterprises will decide whose trust is worth paying for.
