How OpenAI’s Dev Day 2026 Could Influence AI Work

🔍 Read the full analysis: How OpenAI’s Dev Day 2026 Could Influence AI Work on ThorstenMeyerAI.com

Prime Big Deal Days · Oct 6–7Offer from Amazon

Get privacy and security gear delivered free — and shop member deals

• Fast, free delivery on millions of items
• Access to Prime Big Deal Days deals on October 6–7
• Prime Video, Amazon Music and more included

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

OpenAI used its DevDay on 29 September 2026 to announce more than 20 products and features that shift the company’s positioning from chatbot maker to agent platform, with ChatGPT presented as a shared surface for humans and agents that reaches 1.2 billion weekly users. The announcements — including always-on Dots agents, a major plugin upgrade, a Decisions API and Sign in with ChatGPT — carry direct consequences for anyone who builds or sells software, because they change how work gets assigned and how services reach customers.

Five announcements stood out for developers and businesses. Dots are always-on personal agents powered by GPT-6 Astra, each running on its own cloud computer with its own browser and access to more than 4,000 apps through plugins. They are available now to Pro and Business Premium customers in eligible markets, with Enterprise, Edu and Healthcare access via an admin-enabled beta. The first dot is included in the plan; chatting with it does not count toward usage limits, though tasks it starts in Codex or ChatGPT Work do.

Plugins became the central integration layer for ChatGPT, Codex and dots. OpenAI introduced plugin extensions that live inside the ChatGPT sidebar, a new Plugin Creator and submission flow, improved discovery, Sites that host plugins for teams, and support for the proposed MCP Events specification, which lets a plugin trigger an automation when something changes in a connected app. Figma and Adobe showed integrations, and an OpenAI Marketplace lets eligible enterprise customers spend part of their OpenAI commitment with 32 launch partners including Salesforce, ServiceNow, Harvey, CrowdStrike and Baseten.

The Decisions API, in limited preview, applies GPT-6 Luna to narrow, repetitive choices: developers define questions with a finite list of allowed answers and receive a selection usable for classifying content, routing requests or choosing an agent’s next action. The Agents API now supports computer use, letting agents operate software through its interface, with OpenAI running the Codex harness. Sign in with ChatGPT lets users carry their existing plan to 16 launch partners, which OpenAI’s framing suggests makes a free core app with paid upsells commercially workable. On the model side, GPT-6.1 Sol offers near-Astra intelligence at one-fifth of Astra’s token prices.

Why Agent Distribution Changes Software Sales

The core shift for builders is that, in OpenAI’s framing, an agent picks a business by outcome, not by brand. A customer asks for a result, ChatGPT understands the job, and a plugin is selected to do the work — meaning a business can be hired before the customer knows its name. That inverts the usual marketing funnel: the questions become why a plugin was selected or ignored, pushing developers to measure intent, activation and completion rather than impressions or brand awareness.

For enterprises, specialist dots — agents with their own identity, credentials and IT-provisioned hardware — point toward AI taking on defined roles. OpenAI is starting with pilots in procurement, invoice processing, email marketing, customer support and commercial contracting, with a planned Microsoft Agent 365 integration for governance. If those pilots succeed, routine knowledge work could increasingly be delegated rather than performed, affecting hiring, tooling budgets and compliance processes.

The pricing structure also matters: because the first dot is bundled into existing plans and GPT-6.1 Sol sharply undercuts Astra’s token prices, OpenAI is lowering the cost of running agents continuously, which could accelerate adoption among cost-sensitive teams.

From Chatbot to Platform: OpenAI’s Trajectory

OpenAI has held DevDay events annually to court developers, but this year’s edition marks the company’s most explicit move yet from a conversational product to an agent platform. The plugin system announced here replaces and extends earlier integration efforts, unifying how ChatGPT, Codex and the new dots connect to external services through MCP servers.

Signs of the broader strategy appear in the developer documentation, which lists conversion specifications for restaurant reservations, quote requests and product checkout, plus a guide for submitting a Claude Code plugin — suggesting OpenAI wants to accept skills and plugins built for competing agent ecosystems. The 1.2 billion weekly user figure OpenAI cited positions ChatGPT as the distribution surface that plugin developers are effectively bidding to reach.

“An agent picks a business by outcome, not by brand.”

— DevDay commentary slides (per source report)

Unanswered Questions on Reliability and Pricing

Several elements remain unproven or incomplete. The Decisions API is in limited preview, with a broad release promised but not yet dated. Texting with dots is listed as coming, while voice calls work today. OpenAI itself warns that dots can make mistakes, and no independent data yet exists on their accuracy or safety in sustained autonomous use.

Pricing beyond the first dot is also unresolved — OpenAI says users can later add dots or buy more speed and monthly output per dot, but detailed tiers were not announced. It is not yet clear how the enterprise specialist-dot pilots in procurement, contracting and support will perform, nor when the Microsoft Agent 365 governance integration will ship. Marketplace eligibility criteria for enterprise customers were likewise not fully specified.

Rollout Timeline Developers Should Watch

Watch for the Decisions API’s broad release beyond limited preview, expansion of dot access beyond Pro and Business Premium tiers, and the arrival of texting support for dots. The enterprise specialist-dot pilots in procurement, invoice processing, email marketing, customer support and commercial contracting will indicate whether role-specific agents hold up in production.

Developers should monitor the plugin directory’s ranking behavior as discovery improves, the adoption of MCP Events automations, and whether the marketplace’s 32 launch partners expand. The Microsoft Agent 365 integration for governance is a further milestone. For businesses, the practical next step is deciding whether to build a plugin now — since agent-mediated selection may increasingly decide who gets the work.

Key Questions

What are OpenAI’s Dots?

Dots are always-on personal agents powered by GPT-6 Astra, each with its own cloud computer and browser, connecting to over 4,000 apps through plugins. They work toward user goals around the clock, learn preferences over time, and can manage several projects at once.

Who can use Dots right now?

Pro and Business Premium customers in eligible markets have access now. Enterprise, Edu and Healthcare organizations can use a beta once an administrator enables it; it is off by default. The first dot is included in the plan.

What is the Decisions API?

It applies GPT-6 Luna to narrow, repetitive choices. You define questions with a finite list of allowed answers, send text or images as context, and receive a selection your app can use to classify content, route requests or pick an agent’s next action. It is currently in limited preview.

How do the plugin changes affect developers?

Plugins are now the integration layer for ChatGPT, Codex and dots. New features include sidebar extensions, a Plugin Creator with a new submission flow, better discovery, team-hosted Sites, MCP event triggers and an enterprise marketplace — meaning plugin visibility may determine which businesses receive agent-assigned work.

What controls exist over what a Dot can do?

Each dot runs on its own cloud computer, separate from the user’s machine. Custom Rules let users allow, require approval for, or block specific actions, an Activity View shows background work, and monitoring can pause or stop a dot. Sensitive tasks such as changing a password always stay with the user, though OpenAI cautions dots can still make mistakes.

Source: ThorstenMeyerAI.com