{"schemaVersion":1,"editionId":"2026-09-26","title":"Copilot’s next workspace, AI creation tools and the limits of summaries","dek":"Microsoft sets out staged Copilot previews, plus dated catch-ups on Suno MIDI, Gemini avatars, Horizon games, local inference, AI infrastructure and Snapchat.","coverageNote":"Reporting window assumed: September 25, 08:00–September 26, 08:00 CEST, because no completed research cutoff is available. Includes explicitly dated catch-ups from September 17–24. Discovery and the rolling September 12–26 gap scan remain partial; date-only sources cannot establish exact rolling-window inclusion. Product descriptions are documentation-based and not hands-on tested.","summaryIntro":"Today’s selection focuses on what is accessible, what remains a preview, and what the evidence actually measures. Microsoft’s next Copilot features arrive in stages; creative tools come with plan or early-access limits; and a memory experiment shows why checking a summary against its source still matters.","publishedAt":"2026-09-26T06:15:18.357Z","updatedAt":"2026-09-26T09:35:19.368Z","version":2,"items":[{"id":"copilot-home-code-autopilot","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/copilot-home-code-autopilot/43ba7c15955ffa20ee82d085ac02d5ef62b33a69cb0c5483f5b5a256bf073b5d","alt":"Conceptual editorial paper collage: copilot next.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Microsoft outlines Copilot Home, Code and Autopilot—with staged preview access","category":"Tools & Applications","tags":["productivity","coding"],"eventDate":"2026-09-25","status":["announced","company-claim"],"sourceIds":["S1","S2"],"short":"Microsoft announced a Copilot redesign on September 25: Home combines Chat and Cowork, Code builds solutions, and Autopilot handles continuing tasks. Home and Code are due in Frontier over the coming weeks; Autopilot’s private preview expands at month-end. Agentic features carry usage-based billing.","claims":[{"text":"Home and Code are scheduled for Frontier in coming weeks; Autopilot expands to private preview at month-end.","sourceIds":["S1"]},{"text":"Frontier workplace access depends on licensing and organization opt-in.","sourceIds":["S2"]}],"long":[{"text":"Microsoft describes Home as a shared starting point for conversational work and delegated tasks, Code as a way to build applications, and Autopilot as a persistent cloud agent. The feature-specific rollout dates make this an announcement of staged access, rather than a universal launch.","sourceIds":["S1"]},{"text":"For workplace Frontier access, Microsoft lists a Copilot license and organization opt-in. Availability and costs vary by feature, role and subscription; its announcement puts Cowork, Code and Autopilot on usage-based billing.","sourceIds":["S1","S2"]}]},{"id":"suno-studio-midi-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/suno-studio-midi-catchup/5b672f986c71e8a601e4764530f6b26f1b2158d5b9008ec3ebf64653c7f4d089","alt":"Conceptual paper collage of a keyboard and music workflow, with the headline A Better MIDI Workflow.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Suno improves MIDI workflows in its Premier-only Studio","category":"Creativity & Media","tags":["audio"],"eventDate":"2026-09-17","status":["available","company-claim"],"sourceIds":["S3","S4","S5"],"short":"Suno’s September 17 Studio update targets MIDI-to-audio generation and prompting new MIDI clips. The company reports better accuracy and speed, which we have not tested. Studio remains part of paid Premier; Suno recommends Chrome and says Web MIDI is unavailable in Safari.","claims":[{"text":"The September 17 release describes MIDI improvements and clip renaming.","sourceIds":["S3"]},{"text":"Studio requires Premier; Web MIDI is unavailable in Safari.","sourceIds":["S4","S5"]}],"long":[{"text":"The update lets musicians use existing MIDI clips as the basis for generated audio stems, and Suno says chat-generated MIDI now follows requests more reliably. Clips can also be renamed individually or through chat.","sourceIds":["S3"]},{"text":"Current help and pricing pages place Studio in Premier. The browser-based workspace does not support mobile devices; Suno recommends Chrome. This is a dated product update based on documentation, not a hands-on assessment.","sourceIds":["S4","S5"]}]},{"id":"gemini-live-avatar-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/gemini-live-avatar-catchup/6cb429636b5aa2f09d926cc058206fa04a8c39016de6e18eada1bf2e33d4055e","alt":"Conceptual editorial paper collage: a face for live ai.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Gemini Live Avatar reaches enterprise users, with custom avatars allowlisted","category":"Models & Agents","tags":["models","audio","video"],"eventDate":"2026-09-24","status":["available","company-claim"],"sourceIds":["S6","S7"],"short":"Google introduced Gemini 3.8 Live with Live Avatar on September 24 for Gemini Enterprise. It combines conversational speech with generated visual presence and background tool calls. Preset characters are offered, while creating a custom avatar from a reference image requires enterprise allowlisting.","claims":[{"text":"Google announced enterprise Live Avatar availability on September 24 and restricts custom creation to enterprise allowlisting.","sourceIds":["S6"]}],"long":[{"text":"Google’s announcement describes audio-and-video conversations that can continue while tools fetch information in the background. These are vendor-described capabilities, not independently measured latency or realism results.","sourceIds":["S6"]},{"text":"Custom character creation is restricted to enterprise allowlisting. Google’s base-model documentation lists us, eu and us-central1 service regions; that list should not be read as confirmation that every avatar option is enabled for every account. The announcement concerns enterprise access.","sourceIds":["S6","S7"]}]},{"id":"horizon-create-studio-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/horizon-create-studio-catchup/8b9893ae4589511fe20576e09fe98d9b8b4013e9503380b0bdc5701d56f4eac9","alt":"Conceptual editorial paper collage: build a game.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Meta opens a waitlist for Horizon’s mobile and browser game builders","category":"Creativity & Media","tags":["coding","productivity"],"eventDate":"2026-09-24","status":["announced","company-claim"],"sourceIds":["S8"],"short":"Meta’s September 24 Horizon recap describes mobile Create and browser-based Studio for building games with prompts. Projects can move between the two tools. Access is rolling out gradually through a waitlist; publishing makes a game eligible for distribution across Meta platforms, without guaranteeing an audience.","claims":[{"text":"Meta describes gradual early access, mobile/browser project transfer and distribution eligibility.","sourceIds":["S8"]}],"long":[{"text":"Create is designed for phone-based prompting and iteration; Studio adds visual scene editing and more detailed controls in a browser. Meta says both share project assets and history.","sourceIds":["S8"]},{"text":"The tools are being tested with creators, with broader access still ahead. Meta also says games undergo review before publishing and distribution. The announcement does not provide a full price or region list, so the practical next step is checking early-access eligibility.","sourceIds":["S8"]}]},{"id":"liquid-vl-dspark-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/liquid-vl-dspark-catchup/bc3343f2064c78f7eb485059532c72ca8b2b102738e5cf7c5a96500389064343","alt":"Conceptual editorial paper collage: local ai.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Liquid AI releases an experimental accelerator for its vision-language model","category":"Models & Agents","tags":["models"],"eventDate":"2026-09-24","status":["available","company-claim"],"sourceIds":["S9","S10","S11"],"short":"Liquid AI’s September 24 DSpark release adds an experimental draft model for LFM2.5-VL-3B, with llama.cpp, MLX-VLM and SGLang support. It accelerates text decoding; image encoding and prompt processing still take time. The downloadable weights have revenue-based commercial license restrictions.","claims":[{"text":"The release accelerates decoding, leaving vision encoding and prefill unchanged.","sourceIds":["S9"]},{"text":"The supplied license limits commercial use by revenue threshold.","sourceIds":["S11"]}],"long":[{"text":"The smaller model proposes tokens for the main model to verify. Liquid reports faster inference in its own tests, but those measurements used 16-bit processing; quantized-model acceleration was outside the release’s evaluation. No speedup was tested by AI Daywire.","sourceIds":["S9"]},{"text":"The model card marks this as experimental. Its LFM1.0 license contains a commercial-use revenue threshold, despite broad wording in the launch blog about unrestricted deployment. Read those terms before choosing it for a commercial system.","sourceIds":["S10","S11"]}]},{"id":"akamai-anthropic-cpu-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/akamai-anthropic-cpu-catchup/7571015b3e92dc8e9da5607beca400a6fdc34d75dd5a2f3a42a44466a71009f4","alt":"Conceptual editorial paper collage: a major cpu deal.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Akamai announces an $11.6 billion Anthropic CPU agreement","category":"Business & Infrastructure","tags":["business"],"eventDate":"2026-09-24","status":["announced","company-claim"],"sourceIds":["S12"],"short":"Akamai announced a seven-year, $11.6 billion contractual commitment from Anthropic on September 24 for CPU workloads. A further $9 billion is potential expansion, not an additional completed commitment. The deal highlights the general computing capacity needed alongside AI models.","claims":[{"text":"Akamai states an $11.6 billion seven-year CPU commitment, with up to $9 billion in potential expansion.","sourceIds":["S12"]}],"long":[{"text":"The issuer says Anthropic will use Akamai’s distributed cloud infrastructure and software. Akamai also issued a warrant whose continued vesting depends on expansion of the relationship. The announced contract value should not be confused with revenue already earned.","sourceIds":["S12"]},{"text":"For teams planning agent systems, the relevant point is the scope: the stated demand is for CPU workloads. This announcement does not establish a new model capability or a measured improvement for Claude users.","sourceIds":["S12"],"type":"analysis"}]},{"id":"snap-lens-plus-catchup","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/snap-lens-plus-catchup/61d2477a843b8a39a79fadefabb19462d7b17c7c6c634cb862225155600acac3","alt":"Conceptual editorial paper collage: chat to song.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: Snapchat adds chat-to-song and remix tools to paid Lens+","category":"Creativity & Media","tags":["audio","images"],"eventDate":"2026-09-21","status":["available","company-claim"],"sourceIds":["S13"],"short":"Snap’s September 21 Lens+ update adds songs generated from chat messages, photo adjustments, Snap and sticker remixing, and styled captions. These are premium-tier features rolling out across Lens+ markets, with regional availability varying over the following weeks.","claims":[{"text":"Snap announced the creative additions on September 21 with a region-dependent Lens+ rollout.","sourceIds":["S13"]}],"long":[{"text":"Snap says Create Song lets a subscriber choose a chat message and genre, then preview the generated audio before sharing. AI Photoshoot adjusts images, while remix tools offer new treatments for friends’ Snaps and stickers.","sourceIds":["S13"]},{"text":"The announcement covers the paid Lens+ tier, not every Snapchat account. Check whether the feature has reached your region. These descriptions come from Snap’s documentation; AI Daywire has not tested the tools.","sourceIds":["S13"]}]},{"id":"ai-summary-memory-study","kind":"news","thumbnail":{"src":"/api/media/thumbnails/2026-09-26/ai-summary-memory-study/dd21dd0b01e51abd5f31261936a50425fcc88fc8e8aa3e2c1c11ee6d14f2335f","alt":"Conceptual editorial paper collage: check the source.","credit":"AI Daywire · AI-assisted editorial illustration","kind":"editorial-illustration"},"title":"Catch-up: a study finds misleading AI summaries can alter recall of a video","category":"Science & Research","tags":["research","safety"],"eventDate":"2026-09-23","status":["research-preprint"],"sourceIds":["S14","S15"],"short":"A September 23 preprint reports that misleading AI-generated summaries reduced recall accuracy in a controlled traffic-video experiment with 328 analyzed participants. Researchers manipulated a traffic-sign detail in the summaries. The finding supports checking original evidence; it is not a general error rate for AI-assisted work.","claims":[{"text":"The experiment analyzed 328 participants and compared recall after consistent versus misleading summaries.","sourceIds":["S14"]}],"long":[{"text":"Participants watched an animated accident video and later read a summary with a matching or misleading traffic-sign description. Correct sign recall was 83.6% after matching summaries and 44.8% after misleading ones. The summary text was AI-generated, with researchers controlling the critical detail.","sourceIds":["S14"]},{"text":"The study used short animated videos and adult US participants. Its controlled setup does not measure how often ordinary AI summaries cause false memories in daily use. We read the public preprint; the authors also identify a shortened conference version.","sourceIds":["S14","S15"]}]}],"sources":[{"id":"S1","title":"Introducing the new Copilot with Home, Code and Autopilot","publisher":"Microsoft","url":"https://blogs.microsoft.com/blog/2026/09/25/introducing-the-new-copilot-with-home-code-and-autopilot/","publishedDate":"2026-09-25","kind":"vendor-announcement"},{"id":"S2","title":"Explore AI Early Access in Microsoft 365","publisher":"Microsoft","url":"https://www.microsoft.com/en-us/copilot/resources/frontier-program","publishedDate":null,"kind":"official-documentation"},{"id":"S3","title":"Suno Studio: Improved MIDI and more","publisher":"Suno","url":"https://suno.com/release-notes/studio-improved-midi","publishedDate":"2026-09-17","kind":"vendor-release-notes"},{"id":"S4","title":"Introducing Suno Studio 2.0","publisher":"Suno","url":"https://help.suno.com/en/articles/13670529","publishedDate":"2026-08-13","kind":"official-documentation"},{"id":"S5","title":"Suno Pricing","publisher":"Suno","url":"https://suno.com/pricing","publishedDate":null,"kind":"official-pricing"},{"id":"S6","title":"Introducing Gemini 3.8 Live with Live Avatar","publisher":"Google","url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-with-live-avatar/","publishedDate":"2026-09-24","kind":"vendor-announcement"},{"id":"S7","title":"Gemini 3.8 Live","publisher":"Google Cloud","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-live","publishedDate":"2026-09-25","kind":"official-documentation"},{"id":"S8","title":"Meta Connect Recap: Sign Up for Early Access to Horizon Create and Horizon Studio","publisher":"Meta","url":"https://developers.meta.com/blog/meta-connect-recap-horizon-create-and-horizon-studio/","publishedDate":"2026-09-24","kind":"vendor-announcement"},{"id":"S9","title":"LFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyond","publisher":"Liquid AI","url":"https://www.liquid.ai/blog/lfm2-5-vl-dspark","publishedDate":"2026-09-24","kind":"vendor-research-announcement"},{"id":"S10","title":"LFM2.5-VL-3B-DSpark model card","publisher":"Liquid AI","url":"https://huggingface.co/LiquidAI/LFM2.5-VL-3B-DSpark","publishedDate":null,"kind":"model-card"},{"id":"S11","title":"LFM2.5-VL-3B-DSpark LICENSE","publisher":"Liquid AI","url":"https://huggingface.co/LiquidAI/LFM2.5-VL-3B-DSpark/blob/main/LICENSE","publishedDate":null,"kind":"model-license"},{"id":"S12","title":"Akamai Announces $11.6 Billion Multi-year Agreement with Anthropic to Support Growing Demand","publisher":"Akamai","url":"https://www.ir.akamai.com/static-files/e742c1c3-d75a-4fcb-a6f5-df289f242c95","publishedDate":"2026-09-24","kind":"issuer-announcement"},{"id":"S13","title":"Turn Chats Into Songs, Enhance Snaps, Remix Stickers and more with new Features in Lens+","publisher":"Snap","url":"https://newsroom.snap.com/new-lens-plus-ai-features","publishedDate":"2026-09-21","kind":"vendor-announcement"},{"id":"S14","title":"AI-Enabled Human Memory Manipulation: Misleading AI-Generated Summaries Distort Human Memory","publisher":"Mattea Sim, Yael Eiger and Tadayoshi Kohno","url":"https://arxiv.org/html/2609.28820v1","publishedDate":"2026-09-23","kind":"research-preprint"},{"id":"S15","title":"AI-Enabled Human Memory Manipulation — abstract and submission record","publisher":"arXiv","url":"https://arxiv.org/abs/2609.28820","publishedDate":"2026-09-23","kind":"research-preprint"}],"review":{"passed":true,"reviewedAt":"2026-09-26T06:13:25.529Z","method":"source-check"},"corrections":[{"date":"2026-09-26T09:35:15.404Z","note":"Added explicit archive topic tags to all eight stories,  Article text, sources and thumbnails are unchanged."}],"canonicalUrl":"https://aidaywire.com/briefings/2026-09-26","language":"en","aiGenerated":true,"summary":"Today’s selection focuses on what is accessible, what remains a preview, and what the evidence actually measures. Microsoft’s next Copilot features arrive in stages; creative tools come with plan or early-access limits; and a memory experiment shows why checking a summary against its source still matters.\n\nMicrosoft announced a Copilot redesign on September 25: Home combines Chat and Cowork, Code builds solutions, and Autopilot handles continuing tasks. Home and Code are due in Frontier over the coming weeks; Autopilot’s private preview expands at month-end. Agentic features carry usage-based billing.\n\nSuno’s September 17 Studio update targets MIDI-to-audio generation and prompting new MIDI clips. The company reports better accuracy and speed, which we have not tested. Studio remains part of paid Premier; Suno recommends Chrome and says Web MIDI is unavailable in Safari.\n\nGoogle introduced Gemini 3.8 Live with Live Avatar on September 24 for Gemini Enterprise. It combines conversational speech with generated visual presence and background tool calls. Preset characters are offered, while creating a custom avatar from a reference image requires enterprise allowlisting.\n\nMeta’s September 24 Horizon recap describes mobile Create and browser-based Studio for building games with prompts. Projects can move between the two tools. Access is rolling out gradually through a waitlist; publishing makes a game eligible for distribution across Meta platforms, without guaranteeing an audience.\n\nLiquid AI’s September 24 DSpark release adds an experimental draft model for LFM2.5-VL-3B, with llama.cpp, MLX-VLM and SGLang support. It accelerates text decoding; image encoding and prompt processing still take time. The downloadable weights have revenue-based commercial license restrictions.\n\nAkamai announced a seven-year, $11.6 billion contractual commitment from Anthropic on September 24 for CPU workloads. A further $9 billion is potential expansion, not an additional completed commitment. The deal highlights the general computing capacity needed alongside AI models.\n\nSnap’s September 21 Lens+ update adds songs generated from chat messages, photo adjustments, Snap and sticker remixing, and styled captions. These are premium-tier features rolling out across Lens+ markets, with regional availability varying over the following weeks.\n\nA September 23 preprint reports that misleading AI-generated summaries reduced recall accuracy in a controlled traffic-video experiment with 328 analyzed participants. Researchers manipulated a traffic-sign detail in the summaries. The finding supports checking original evidence; it is not a general error rate for AI-assisted work.","fullText":"Microsoft outlines Copilot Home, Code and Autopilot—with staged preview access\n\nMicrosoft describes Home as a shared starting point for conversational work and delegated tasks, Code as a way to build applications, and Autopilot as a persistent cloud agent. The feature-specific rollout dates make this an announcement of staged access, rather than a universal launch.\n\nFor workplace Frontier access, Microsoft lists a Copilot license and organization opt-in. Availability and costs vary by feature, role and subscription; its announcement puts Cowork, Code and Autopilot on usage-based billing.\n\nCatch-up: Suno improves MIDI workflows in its Premier-only Studio\n\nThe update lets musicians use existing MIDI clips as the basis for generated audio stems, and Suno says chat-generated MIDI now follows requests more reliably. Clips can also be renamed individually or through chat.\n\nCurrent help and pricing pages place Studio in Premier. The browser-based workspace does not support mobile devices; Suno recommends Chrome. This is a dated product update based on documentation, not a hands-on assessment.\n\nCatch-up: Gemini Live Avatar reaches enterprise users, with custom avatars allowlisted\n\nGoogle’s announcement describes audio-and-video conversations that can continue while tools fetch information in the background. These are vendor-described capabilities, not independently measured latency or realism results.\n\nCustom character creation is restricted to enterprise allowlisting. Google’s base-model documentation lists us, eu and us-central1 service regions; that list should not be read as confirmation that every avatar option is enabled for every account. The announcement concerns enterprise access.\n\nCatch-up: Meta opens a waitlist for Horizon’s mobile and browser game builders\n\nCreate is designed for phone-based prompting and iteration; Studio adds visual scene editing and more detailed controls in a browser. Meta says both share project assets and history.\n\nThe tools are being tested with creators, with broader access still ahead. Meta also says games undergo review before publishing and distribution. The announcement does not provide a full price or region list, so the practical next step is checking early-access eligibility.\n\nCatch-up: Liquid AI releases an experimental accelerator for its vision-language model\n\nThe smaller model proposes tokens for the main model to verify. Liquid reports faster inference in its own tests, but those measurements used 16-bit processing; quantized-model acceleration was outside the release’s evaluation. No speedup was tested by AI Daywire.\n\nThe model card marks this as experimental. Its LFM1.0 license contains a commercial-use revenue threshold, despite broad wording in the launch blog about unrestricted deployment. Read those terms before choosing it for a commercial system.\n\nCatch-up: Akamai announces an $11.6 billion Anthropic CPU agreement\n\nThe issuer says Anthropic will use Akamai’s distributed cloud infrastructure and software. Akamai also issued a warrant whose continued vesting depends on expansion of the relationship. The announced contract value should not be confused with revenue already earned.\n\nFor teams planning agent systems, the relevant point is the scope: the stated demand is for CPU workloads. This announcement does not establish a new model capability or a measured improvement for Claude users.\n\nCatch-up: Snapchat adds chat-to-song and remix tools to paid Lens+\n\nSnap says Create Song lets a subscriber choose a chat message and genre, then preview the generated audio before sharing. AI Photoshoot adjusts images, while remix tools offer new treatments for friends’ Snaps and stickers.\n\nThe announcement covers the paid Lens+ tier, not every Snapchat account. Check whether the feature has reached your region. These descriptions come from Snap’s documentation; AI Daywire has not tested the tools.\n\nCatch-up: a study finds misleading AI summaries can alter recall of a video\n\nParticipants watched an animated accident video and later read a summary with a matching or misleading traffic-sign description. Correct sign recall was 83.6% after matching summaries and 44.8% after misleading ones. The summary text was AI-generated, with researchers controlling the critical detail.\n\nThe study used short animated videos and adult US participants. Its controlled setup does not measure how often ordinary AI summaries cause false memories in daily use. We read the public preprint; the authors also identify a shortened conference version.","checksum":"b68f38f5f63f9c87fe74d997cfa82cad45c0270db6c08db6f3314920340856e4"}