Worldcoin (WLD) Co-Founder Altman's OpenAI Launches Dots Agents Across 4,000 Apps

DevDay Reveal Answers Meta's Muse

OpenAI moved into always-on artificial intelligence agents on September 29, unveiling Dots at its DevDay keynote in what the industry immediately read as a direct answer to Meta's Muse. Meta's agent, introduced on September 8 through Meta's Muse announcement, got off to a fast start: data from Apptopia shows 1.8 million downloads in its first 12 days, ahead of the 1.3 million ChatGPT recorded over the same window. The two products price differently from the ground up — Muse runs on a free base tier with paid subscriptions beyond usage limits, while Dots ships inside OpenAI's paid plans, giving Pro and Business Premium subscribers one dot at no additional cost. Geography splits too: Muse is limited to the United States and Canada, whereas Dots has reached Pro users in markets outside the European Economic Area, Switzerland and the United Kingdom, including Taiwan. Architecturally, the rivals converge on the same design — each user's agent lives on a dedicated, long-running cloud virtual machine, so assigned tasks keep executing after the app is closed. Where they part ways is trust. Muse drew criticism after it reportedly synced a user's iMessage despite an explicit authorization refusal, and Amazon blocked the agent over unauthorized-access concerns; Meta's small-business build now stresses that nothing is published, sent or spent without user consent. OpenAI, learning from those stumbles, spent much of its launch briefing on permission controls. Chief executive Sam Altman — who co-founded Worldcoin and whose company is pursuing a reported $30 billion raise at a $1.4 trillion valuation — called Muse “a good-looking product” while insisting he feels no threat from Mark Zuckerberg and remains confident in OpenAI's roadmap. xAI's Grok Bot entered the same lane earlier, in August, bundled with SuperGrok plans from $30 a month.

4,000-App Reach and a $500 Pro Plan

What separates a dot from a chatbot is persistence: it works toward user goals continuously, without daily supervision, across a network of connected software. Each dot runs on GPT-6 Astra — the model family whose base GPT-6 pricing OpenAI recently halved by 50% — paired with its own cloud computer and browser, and it can write and test code when a task demands it; with explicit permission it can attach to a personal machine, though local access stays switched off by default. Reach is the headline: more than 4,000 connected applications, spanning workplace software and DApps alike, plus collaboration inside ChatGPT, Slack and Teams, with SMS access for Pro users already in a limited United States trial. Enterprise governance runs through Microsoft's Agent 365, letting firms manage dedicated dots with existing identity and access controls. Deployment began September 29: eligible Pro and Business Premium subscribers receive one dot free, Enterprise customers get admin-activated beta access, and OpenAI's announcement sets Business Premium at $100 per user monthly on annual billing, or $125 month-to-month, alongside a new $500-per-month Pro plan unveiled at DevDay. Pro access has not launched in the EEA, Switzerland or the UK. The product extends Frontier, the February framework for agents operating across a company's apps, data and tools — and lands where the revenue is: business clients already contribute roughly 40% of OpenAI's income, a share CFO Sarah Friar projects could reach 50% by the end of 2026. The stakes surfaced a day early, when OpenAI withdrew GPT-6.1 Astra after internal safety evaluations fell short, a reminder that agent reliability remains an unsolved industry problem. Forecasts add context: Gartner pegs 2026 global AI platform and model spending at $64.25 billion, up 63.4% from $39.31 billion in 2025, and Capgemini sizes the agent opportunity at $450 billion by 2028 — although only 2% of organizations currently deploy agents at scale. Readers tracking the market in real time can follow live spot and futures prices on Bitget.

Why Worldcoin (WLD) Is Watching

For Worldcoin (WLD), the through-line is Sam Altman's double role. The Worldcoin protocol, built by Tools for Humanity, targets the exact problem agentic platforms now face at scale: proving that a human — not an agent — stands behind an account, an inbox or a payment, as software increasingly writes code, sends mail and spends money, a scenario adjacent to the DeFAI thesis of autonomous on-chain finance. Worldcoin ecosystem observers watched Altman this year press the White House on delaying UK AI tests and urge UN oversight of AI decisions. OpenAI's own launch material describes agents acting across connected apps without daily supervision — our desk reads every expansion of that autonomy as strengthening the demand case for verified personhood, and for World ID.