Microsoft's MAI-Transcribe-2 covers 60 languages with speaker labels and word timestamps, though streaming is not among its documented features. Live voice products get a fast model for replies and still need another way to hear the caller.
Perspective Coverage
8 publishers
- Builder
- Builder 53%
- Operator
- Operator 30%
- Investor
- Investor 17%
Reality
- Evidence35
- Adoption15
- Hype gap−35
- Incentives65
- Confidence70
LiveKit Agents now speaks in Microsoft's MAI voices, including MAI-Voice-2-Flash, through an official plugin that covers text-to-speech only. Teams no longer write their own adapter, but every deployment still needs an Azure Speech key and region.
Reality
- Evidence45
- Adoption
- Insufficient
- Hype gap+5
- Incentives45
- Confidence50
Cohere's Embed 5 lets teams index with Pro at $0.12 per million text tokens and query with Fast at $0.08 against the same vectors. The quality and throughput figures behind that split come from Cohere's own tests, run on datasets and parsing pipelines a buyer may not share.
Reality
- Evidence45
- Adoption
- Insufficient
- Hype gap+20
- Incentives75
- Confidence55
Cohesity's Agent Resilience rolls Bedrock AI agents back to immutable recovery points, yet its product leads say emails already sent stay sent. Outside systems still need their own clean-up and tests, and the company calls its recovery-time edge intended, not measured.
Reality
- Evidence35
- Adoption
- Insufficient
- Hype gap0
- Incentives70
- Confidence40
Microsoft is splitting Copilot into Home, Code and Autopilot, with its always-on Autopilot agent entering private preview in late September. Autopilot agents carry their own identities and bill by usage, so tenant admins need access reviews and spend plans ready before the first instances arrive.
Perspective Coverage
5 publishers
- Builder
- Builder 33%
- Operator
- Operator 39%
- Investor
- Investor 28%
Reality
- Evidence60
- Adoption
- Insufficient
- Hype gap+35
- Incentives65
- Confidence60
Greg Brockman said AGI arrived with GPT-6 Astra on September 3. The enforcement the launch actually documents is a misuse classifier running inside AWS's service boundary, plus a voluntary 30-day US review that carried no license.
Perspective Coverage
18 publishers
- Builder
- Builder 40%
- Operator
- Operator 34%
- Investor
- Investor 26%
Reality
- Evidence60
- Adoption50
- Hype gap+45
- Incentives78
- Confidence66
Existing subscribers keep their access, but anyone on Free, Go, Plus or the $100 Pro plan is locked out of an upgrade, and a downgrade taken during the pause cannot be undone until OpenAI reopens sales.
Perspective Coverage
4 publishers
- Builder
- Builder 25%
- Operator
- Operator 50%
- Investor
- Investor 25%
Reality
- Evidence72
- Adoption45
- Hype gap+25
- Incentives60
- Confidence68
OpenAI released GPT-6 Astra on September 3, and by September 14 it was generally available on Amazon Bedrock with a million-token input window. Apollo Research had three days with a near-final build.
Perspective Coverage
4 publishers
- Builder
- Builder 36%
- Operator
- Operator 49%
- Investor
- Investor 15%
Reality
- Evidence55
- Adoption25
- Hype gap+30
- Incentives60
- Confidence50
Microsoft says evaluations should decide which GPT-6 model runs each agent. The serving matrix leaves Luna without reserved capacity. GitHub turns both new models on for admins who left the default alone.
Publishers:azure.microsoft.com · github.blog Reality
- Evidence56
- Adoption34
- Hype gap+30
- Incentives92
- Confidence68
Microsoft's public preview flags jailbreaks, prompt injection and credential leakage from agent telemetry, and agents built outside Copilot Studio, Foundry and Agent Builder need the Agent 365 SDK first.
Publishers:learn.microsoft.com
Reality
- Evidence52
- Adoption12
- Hype gap+15
- Incentives78
- Confidence58
Claude Code started reading AGENTS.md on September 18, but only where no CLAUDE.md sits in the working directory or any parent above it. Monorepos need a stub file per package or a change to the instructionFiles setting.
Reality
- Evidence55
- Adoption40
- Hype gap+10
- Incentives35
- Confidence55
Microsoft Foundry writes agent runs as OpenTelemetry spans with gen_ai attributes and exports them to Azure Monitor, where evaluators score sampled production traffic off the same records. Those attributes include tool arguments.
Reality
- Evidence40
- Adoption
- Insufficient
- Hype gap+10
- Incentives55
- Confidence45
Microsoft published its retrosynthesis model RetroChimera in Nature and released the weights. The blind test behind the headline claim graded individual disconnections, a narrower question than whether a whole route survives a lab.
Reality
- Evidence62
- Adoption28
- Hype gap+18
- Incentives68
- Confidence55
Version 2.1.277 reads a shared AGENTS.md when no CLAUDE.md is in scope. The documented exceptions cover first sessions, telemetry-disabled runs, Bedrock, Vertex and Foundry, and the hooks governance tooling watches.
Reality
- Evidence68
- Adoption58
- Hype gap+12
- Incentives62
- Confidence64
A Microsoft case study credits Docusign's swap to small task-specific models with 90% lower cost and eight times the throughput. Docusign's own accounting of the whole pipeline claims 50 times cheaper per document.
Reality
- Evidence42
- Adoption62
- Hype gap+24
- Incentives80
- Confidence55
A dev.to walkthrough of Microsoft Foundry's Agent Service pulls the tool declaration apart from the lazily provisioned container and the hour-long billable session. Which of the three you hold a reference to decides whether the second question still has the data.
Reality
- Evidence30
- Adoption
- Insufficient
- Hype gap+20
- Incentives60
- Confidence35
Microsoft treats speech as its own agent kind, with a managed realtime orchestrator holding the socket open. A dev.to walkthrough sets out why turn detection and tool calls had to be restructured around a few hundred milliseconds.
Reality
- Evidence38
- Adoption
- Insufficient
- Hype gap+15
- Incentives42
- Confidence40
Zero-data-retention organizations that want Anthropic's covered models will have to turn retention on workspace by workspace from June 9, 2026, with prompts and outputs held for 30 days for misuse analysis.
Publishers:privacy.claude.com
Reality
- Evidence63
- Adoption
- Insufficient
- Hype gap+12
- Incentives78
- Confidence55
The early-bird rate on MAI-Transcribe-2 sits 72% below the line's April price, which saves a 100,000-hour buyer $26,000 a year and takes 26 cents out of every 36 a standalone vendor charged for the same hour.
Reality
- Evidence44
- Adoption21
- Hype gap+33
- Incentives74
- Confidence45
The mark is applied during generation rather than at an interface, so anything built on the Claude Platform API returns marked text by default. Reading the mark stays behind a private detection API with a registration list.
Reality
- Evidence34
- Adoption28
- Hype gap+14
- Incentives76
- Confidence36
Earlier coverage
- Fable 5.1 binds each thinking block to the exact bytes of the prefix that produced it
Build · September 1, 2026 · 1 publisher
- Per-PTU throughput spans 25x across three models in the same GPT-5.6 family
Build · August 30, 2026 · 1 publisher
- Grok 4.6 reaches Foundry, and xAI's fourth catalog in 14 days turns model choice into a purchase order
Build · August 26, 2026 · 1 publisher
- Same-day GPT-5.6 on Azure kills the parity argument, leaving auth and residency to decide
Build · August 25, 2026 · 1 publisher
- Claude Skills leaves beta at /v1/skills, and the release discipline stays yours
Build · August 24, 2026 · 1 publisher
- Three clouds, one protocol, no stored credentials: the runtime becomes a swappable part
Build · August 23, 2026 · 1 publisher
- Canva's forecast cut turns model routing into a product line item
Product · August 22, 2026 · 1 publisher
- Microsoft puts AI agents in Entra, which makes agent sprawl an identity team problem
Leadership · August 20, 2026 · 1 publisher
- Microsoft's MAI-Thinking-1 lands in Foundry, and .NET teams get a reasoning model without Python
Build · August 18, 2026 · 1 publisher
- Three permission problems wearing one service principal: why published agents return 403
Build · August 15, 2026 · 1 publisher