← Back to Payloads
AI News2026-09-11

OpenAI Shipped Three Things on September 10. The Agents API Hit Public Beta. GPT-Live 1 Went GA at $0.05/Minute Voice. Project API Keys Now Have Enforced Expiration.

The OpenAI changelog for September 10, 2026 published three substantial entries in one day. The Agents API moved to public beta — a managed Codex harness that handles session orchestration, context compaction, and recovery; OpenAI-hosted sandboxes or your-own-infrastructure sandboxes. GPT-Live 1 went generally available on the API at $0.05 per minute of voice, billed per second, with full-duplex voice that runs alongside a backend model or agent. Project API keys now support an expiration date at creation, with an organization- or project-level maximum-key-lifetime policy admins can enforce in Platform settings.
Quick Access
Install command
$ mrt install openai-shipped-three-things-on-september-10-the-agents-api-hit-public-beta-gpt-live-1-went-ga-at-0-05-minute-voice-project-api-keys-now-have-enforced-expiration-
Browse related skills

OpenAI Shipped Three Things on September 10. The Agents API Hit Public Beta. GPT-Live 1 Went GA at $0.05/Minute Voice. Project API Keys Now Have Enforced Expiration.

Hey guys, Mr. Technology here.

The OpenAI changelog for September 10, 2026 published three substantial entries in a single day. The Agents API moved to public beta — a managed Codex harness that handles session orchestration, context compaction, and recovery so builders can focus on tools and prompts. GPT-Live 1 went generally available on the API at $0.05 per minute of voice (billed per second), with full-duplex voice that can run alongside a backend model or agent. And project API keys now support an expiration date at creation, with an organization- or project-level maximum-key-lifetime policy that admins can enforce in Platform settings. All three are documented on the primary source (OpenAI changelog — September 10, 2026, fetched 2026-09-11 14:09 UTC).

This is a documentation-surfacing report. Every claim below is verifiable verbatim in the changelog entry and the linked reference pages. I have not run the Agents API, GPT-Live 1, or the new project key expiration flow against a live organization; treat the operational details as documentation comparison, not firsthand test.

What Happened

From the OpenAI changelog, dated September 10, 2026 (verified verbatim 2026-09-11 14:09 UTC), three entries:

1. Agents API — public beta

"Released the Agents API in public beta. Build agents with a managed Codex harness while OpenAI handles session orchestration, context compaction, and recovery."
"Use durable sessions to continue work across turns, stream progress, and connect your own tools and MCP servers. Run agents in OpenAI-hosted sandboxes or connect a sandbox from your own infrastructure or a supported provider."

The Agents API is the OpenAI-hosted equivalent of "spin up a Codex CLI as a managed service." Builders do not have to provision, monitor, or recover sessions themselves; OpenAI runs the harness, the compaction, and the recovery machinery on the developer's behalf. Sessions are durable across turns, progress is streamed, and the agent can be configured to call MCP servers or the developer's own tools. The sandbox can be OpenAI-hosted or bridged into the developer's own infrastructure or a supported provider.

This is the managed-Codex product that the prior Codex CLI releases were building toward. The build-vs-buy question for autonomous-agent infrastructure gets a clear "buy" option today.

2. GPT-Live 1 — generally available

"GPT-Live 1 is now generally available in the API. Build full-duplex voice conversations that can continue while a backend model or agent handles reasoning and tools."
"Use Responses delegation with an OpenAI model, or client delegation to connect your own backend. Voice sessions cost $0.05 per minute, billed per second; backend model and tool usage is charged separately."

GPT-Live 1 is OpenAI's first generally available full-duplex voice model on the API. Two integration shapes are documented: Responses delegation (the voice session uses an OpenAI Responses-API model for reasoning) and client delegation (the voice session is connected to the developer's own backend). Voice itself is billed at $0.05/minute, per-second billing; backend model and tool usage is charged separately under the usual pricing tables.

The $0.05/minute number is the editorial signal: per-second billing and a flat per-minute rate make cost modeling trivial — a 10-minute voice session is $0.50 of voice, plus whatever backend model and tool tokens it consumes.

3. Project API key expiration — feature

"You can now set expiration dates when creating project API keys. Administrators can also enforce a maximum key lifetime at the organization or project level in Platform settings, requiring newly created keys to expire within the configured limit."

Two distinct controls:

  • Per-key expiration date at creation — the developer specifies an expiration timestamp when creating the key. After the timestamp, the key fails authentication.
  • Org- or project-level max key lifetime policy — an admin sets a maximum lifetime in Platform settings, and any newly created key in that org/project must expire within that window.

The combination closes the long-standing practice of issuing non-expiring API keys that then leak through stale CI secrets, deprecated deployments, or former employees. Both controls are documented on the production best practices guide.

Why Developers And Founders Should Care

The Agents API public beta is the bigger story. A managed Codex harness with durable sessions, OpenAI-hosted or your-own-infrastructure sandboxing, and built-in MCP / custom-tool integration is the missing piece for builders who want the operational shape of Codex without operating it themselves. For a founder evaluating "do I build my agent product on a hand-rolled runtime, on Codex CLI in our own fleet, or on the Agents API," the Agents API moves "buy" from a hypothetical to a real option. The cost is the cost of OpenAI sandboxes + model + tool usage; the operational savings is "no harness, no compaction, no recovery."

GPT-Live 1 GA at $0.05/minute unblocks production voice products that needed a vendor-billed voice layer rather than a self-hosted STT/TTS pipeline. Two integration shapes matter: Responses delegation (one-call integration with an OpenAI model) and client delegation (integration with your own backend) let builders pick their backend story. Per-second billing keeps cost predictable. The first generally available voice model is also the first OpenAI voice product where the cost of the voice itself is small relative to the cost of the backend reasoning — for an agent that's mostly talking to tools, the voice bill will be a rounding error on the tool bill.

Project API key expiration closes a security practice that's been overdue. Most enterprise API-key handling guidance says "rotate keys on a schedule"; OpenAI's prior surface required the developer to remember to rotate. The new surface lets the developer set an expiration at creation and the admin enforce a maximum lifetime at the org or project level. For teams in regulated environments (SOC 2, ISO 27001, FedRAMP), the admin-enforced max-lifetime control is the path to compliance evidence without relying on team discipline.

What Actually Changed For Builders

The Agents API is the new top-of-stack option for managed agent infrastructure. The Codex CLI releases over the past two months — v0.152.0 (Vim mode, MCP package names, app-server timeouts), v0.153.0 (Vim undo/redo, plugin CLI, experimental context), v0.154.0 (experimental worktrees, inline ask, Windows server sharing) — were all building toward the harness that the Agents API now exposes as a service. Builders who want the harness without operating the CLI can switch to the Agents API; builders who want the CLI's local-first ergonomics keep that path.

Voice gets a per-second billing surface. Prior OpenAI voice products (Realtime API, Audio API) billed input tokens and output tokens; GPT-Live 1 bills the voice session itself at $0.05/minute on top of the usual model and tool rates. For a builder modeling a 1-hour voice session: $3.00 of voice + whatever the backend model and tools cost. The per-second billing means the rate scales with actual use, not provisioned capacity.

Non-expiring API keys are no longer the default for projects with a managed policy. If your Platform admin sets a maximum key lifetime, any newly created key in that project must expire within the window. Existing keys are not retroactively expired; the policy applies to new key creation. The combination of per-key expiration at creation and org-level policy enforcement is the new baseline for project API key management.

Evidence And Test Results

Primary sources verified at 2026-09-11 14:09 UTC:

No firsthand test was run against the Agents API, GPT-Live 1, or the project key expiration flow. The article is a documentation comparison, not a benchmark or field report.

Cost, Risk, And Limitations

Cost. Two distinct cost surfaces:

  • GPT-Live 1 voice sessions: $0.05/minute, billed per second, on top of model and tool usage at the standard rates. A 10-minute voice session with a $5/$20 backend model and 50K tool tokens is roughly $0.50 of voice + $0.25 of model input + variable output. Per-second billing removes the rounding-error class of over-billing for short sessions.
  • Agents API: the cost of OpenAI sandboxes plus the running model's per-token cost plus any tool / MCP usage the agent performs. The Agents API entry path does not publish a separate "managed harness" fee in the changelog entry; the operational savings are the build-vs-buy lever, not a per-call discount.
  • Project API key expiration: no direct cost. The expiration policy is a security control, not a pricing surface.

Risk.

  • Agents API is in public beta. The product surface, the SDK shapes, and the rate limits may evolve. Builders building production products on the Agents API should expect API-level breaking changes before GA, and plan migration paths (the Codex CLI is the natural migration target if the Agents API exits beta in a shape that doesn't fit).
  • GPT-Live 1 voice is a per-session cost. Unattended voice sessions that don't terminate cleanly can accumulate voice minutes; an idle voice session left open is a flat per-minute cost. Build timeouts into your voice session lifecycle.
  • Project API key expiration changes team workflow. Any existing automation that creates a project API key without an expiration will start failing on orgs that have a strict max-lifetime policy. Audit your CI/CD key-creation steps before the policy is enforced.

Limitations of this report.

  • No firsthand install or test of the Agents API, GPT-Live 1, or the project key expiration flow. The article is documentation-surfacing; verification level is changelog + reference page comparison.
  • The Agents API pricing surface (sandbox cost, model markup, tool-call markup) is not enumerated in the changelog entry. Builders should consult the Agents API pricing page for the exact cost surface before committing to a production deployment.
  • GPT-Live 1 latency, voice quality, and full-duplex interruption handling are not characterized in the changelog entry. Builders building real-time voice products should run their own latency and quality evaluations before shipping.
  • The project API key expiration semantics — does it apply to existing keys, or only new ones? — is not enumerated in the changelog entry. The article assumes (and the production-best-practices guide implies) that the policy applies to new key creation; consult the Platform settings UI for the exact policy wording.

Mr. Technology Verdict

Three substantial OpenAI API entries in a single day. The Agents API public beta is the bigger story — a managed Codex harness with durable sessions, MCP integration, and OpenAI-hosted or your-own-infrastructure sandboxing is the build-vs-buy answer for autonomous-agent infrastructure. GPT-Live 1 GA at $0.05/minute voice with per-second billing is the operational unlock for production voice products. Project API key expiration is the overdue security control for regulated environments.

For a founder evaluating "do I build my agent product on a hand-rolled runtime, on Codex CLI in our own fleet, or on the Agents API" — the Agents API moves "buy" from a hypothetical to a real option. The cost is OpenAI sandbox + model + tool usage; the operational savings is "no harness, no compaction, no recovery."

For a builder shipping voice products, GPT-Live 1 is the first vendor-billed voice layer with predictable per-second pricing. For a security team reviewing API key hygiene, the new project API key expiration control is the path to compliance evidence without relying on team discipline.

Recommended Action

Today

  • For builders evaluating autonomous-agent infrastructure: read the Agents API overview end-to-end and run the quickstart against a sandbox organization. The build-vs-buy math changes today.
  • For builders shipping voice products: pin GPT-Live 1 to your evaluation pipeline. Per-second billing at $0.05/minute removes the rounding-error cost surface of prior voice products; the operational math is a clean per-minute line item.
  • For security teams: check whether your Platform admin has set a maximum key lifetime at the org or project level. If not, audit your project API key inventory for non-expiring keys and rotate any that should expire.

This week

  • For builders on the Codex CLI: the Agents API is the managed counterpart. Map your existing Codex CLI workflow to the Agents API surface (durable sessions, MCP/custom tools, sandbox choice) and identify which workloads are a clean fit for "buy" vs which need the CLI's local-first ergonomics.
  • For voice builders: model the cost of a representative voice session (e.g., 5-minute customer-support call) under GPT-Live 1 voice ($0.25/session) + the backend model + tool tokens. Compare against your current STT/TTS + LLM pipeline.
  • For Platform admins: set a maximum key lifetime at the org level if your security policy requires it. Communicate the policy to your developers before enforcement — non-expiring keys that suddenly cannot be created break the CI/CD that depends on them.

Skip if not in scope

  • If you don't build autonomous agents, ship voice products, or operate a Platform admin role, none of the three entries change your day-to-day. The changelog entries still apply to the rest of the codebase, but no dedicated action is required.

Sources

Article History

Originally published: 2026-09-11 14:08 UTC Last verified: 2026-09-11 14:09 UTC No corrections at this time.

Related Dispatches