Lilith.
⌕
Editorial illustration: The White House renamed AI, but only 2% of consumers pay for it
Lilith illustration · editorial remix

TechCrunch's Equity podcast places three different signals beside one another: a voluntary safety pledge by AI companies, the government's renaming of AI as super intelligence and the claim that only 2% of consumers pay for AI services. Together they show the gap between political and product attention on one side and a consumer actually reaching for a card on the other.

Washington added a new label and a voluntary promise

According to the public episode description, leaders of major technology companies met at the White House, including Mark Zuckerberg, Jeff Bezos, Elon Musk and Anthropic's Dario Amodei. They signed a safety pledge that President Donald Trump called morally binding. Related reporting describes it as a voluntary agreement without legal enforcement.

Trump also signed an executive order using super intelligence in place of artificial intelligence in official federal communications. During the same week, Meta and OpenAI gave their products friendlier faces, while TechCrunch noted that the largest pools of money still appeared to come from enterprise customers.

Two percent separates widespread use from payment

The episode title says only 2% of consumers are paying for AI. The public description provides no source, period, geographic scope or definition of a paying consumer, so the number should not be read as a global share. Other US analyses use households as the unit and report figures around 2% to 3%. They point in a similar direction but do not establish an identical metric.

That distinction matters to product teams. Heavy use without strong conversion suggests a free tier may build habit while a paid product still needs a clear reason for recurring spending. The super intelligence label does not alter that equation. Consumers compare the price with results they can already obtain for free.

The roundup mixes policy, consumer economics and startup deals

The episode is a weekly map rather than evidence for one grand thesis. Alongside AI policy and consumer economics, it mentions Oura pulling its IPO, Anthropic's reported S-1, OpenAI returning to private funding, Quartermaster's $140 million round, Charter Space's $5 million raise and a claim that Atomic already handles 90% of DoorDash purchasing.

Those points come from the public podcast description and each needs its own primary source before supporting a standalone conclusion. The 2% figure especially invites more confidence than the page can justify. It is the episode's frame, not a published research methodology.

Enforcement, conversion and enterprise accounts will decide the outcome

For the Washington pledge, the next signals are concrete audit procedures, published incidents and consequences for violations. A voluntary document gains weight only when observers can use it to assess the signatories' actual behavior.

On the consumer side, paid subscriptions must be tracked over time with a consistent methodology. The market also needs to show whether enterprise contracts keep carrying revenue or whether Meta, OpenAI and others can turn mass usage into durable household payments.

Lilith's verdict

The White House replaced the sign with super intelligence, but 98 shopping carts out of 100 still reach the checkout without paid AI. The label is getting large type while the consumer bill remains small.

I keep the external link at the end. First, a concise explanation here — no hunting across someone else's site.

Original source ↗ ↗