Dropbox status: access issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Dropbox reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Dropbox users through our website.

• Errors (50%)
• Sign in (33%)
• Website Down (17%)

Live Outage Map

The most recent Dropbox outage reports came from the following cities:

City	Problem Type	Report Time
Bournemouth	Sign in	1 month ago
Paramaribo	Errors	2 months ago
Bogotá	Website Down	2 months ago
Auxerre	Errors	2 months ago
Salt Lake City	Sign in	2 months ago
Madrid	Errors	3 months ago

Full Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

Dropbox Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Harshita Renee (@harshitaxmars) reported 6 minutes ago

  Despite me having proven him wrong about the exact requirement table issue he pinned on me as a “it’s her problem, shut her up” (Dropbox has the scoresheet proving I was not out of line, they were), I don’t think he can ever be wrong. That is just technical error on his part.

• 
  AI Crave (@wecraveai) reported 12 minutes ago

  Open source NotebookLM alternative with no data limits and AI agents. Same idea as Google's NotebookLM. Same chat-with-your-docs. Same podcast generator. Same cited answers. Except this one has no source limit, no notebook limit, no 200MB file cap, and no Google login. It's called SurfSense. Google NotebookLM vs SurfSense: - Sources per notebook: 50 to 600 → Unlimited - File size cap: 200MB and 500K words → No limit - LLM choice: Gemini only → 100+ models via LiteLLM - Local LLMs: Not allowed → Full Ollama and vLLM support - Self-host: No → Yes, one Docker command - Price: $0, $19.99/mo Pro, or $249.99/mo Ultra → $0 forever Here's the wildest part: It connects to 27+ sources Google can't touch. Notion. Slack. Linear. Jira. GitHub. Discord. Dropbox. OneDrive. Gmail. Confluence. Obsidian. ClickUp. Microsoft Teams. Airtable. Your entire work life, indexed once, searchable from one chat box. 14.4K GitHub stars. 1.4K forks. 6,232 commits. Apache-2.0 license. One honest note: the README says it's not yet production-ready and still being actively developed. But it already does more than NotebookLM does, and the gap is widening every release. This is what NotebookLM should have been from the start. Repo in the first comment.

• 
  Adeyinka Prime™ (@adefilaadeyinka) reported 14 minutes ago

  @aarondfrancis @Shpigford Exactly - when sharing solves a problem for the person sharing, it doesn't feel like marketing. Dropbox nailed this because storing files alone was less useful than storing them with others. The product itself created the reason to invite.

• 
  Karus Daedlyn *Cat Dad Era* (@CDaedlyn) reported 15 minutes ago

  @CantEverDie Keep making burner Gmail accounts and sign up for the free stuff. I jest, because that's terrible practice, although I did know an attorney IRL that would do that to daisy-chain those Dropbox sites. Infinite free storage, just in 2.5GB chunks.

• 
  Arthur "lynch mob" van Pelt 🔥 ∞/21M ⚡ (@Arthur_van_Pelt) reported 18 minutes ago

  @laz1m0v @adam3us You want a multifunctional database. Go try BSV or Microsoft SQL Server or Arweave. They have what you want. Bitcoin is money. Not a Pepe Dropbox. Go take your Pepes elsewhere, please. Then we wouldn't even need filters if everyone behaved in the spirit of Cash instead of Pepe.

• 
  Jack (@jacklandas) reported 18 minutes ago

  alibaba banning claude code over backdoor fears is the new "no dropbox on company laptops" every big corp security team is about to have a list. cursor approved, claude code not. windsurf maybe. this fragmentation is going to be a real problem for devs who just want to ship

• 
  Paulius Eidukas (艾文) ☭ (@nilsenist) reported 18 minutes ago

  @RavenT1me Sorry about that! The hacked account asks to download a virus disguised as an "indie game". I've reported that both to Discord and Dropbox/GitHub/YouTube which help distribute the file. Hopefully that shuts down the hacker at least temporarily. Hope you get your account back!

• 
  Abhishek Singh (@0xlelouch_) reported 20 minutes ago

  Interviewer: design Dropbox file sync. I paused and asked what they meant by sync. Whole product? Or just the client protocol? Single user? Team shares? Offline edits? Large files? Mobile on spotty networks? End to end encryption? What’s the SLO for conflict rate and time to converge? Once we scoped it to single-user sync across devices with offline support, I wrote requirements: detect changes, upload deltas, download updates, handle conflicts, resumable transfers, and don’t melt the battery. Non-goals: shared folders and fine-grained permissions. APIs and data model next. I used a file ID stable across renames, plus per-file version and per-device cursor. Client calls: /changes?cursor=..., /upload_session/start, /upload_session/append, /upload_session/commit, /download?file_id&version, /ack?cursor. Server tables: file_metadata(file_id, user_id, path, type, size, content_hash, current_version), file_versions(file_id, version, blob_ref, created_at), device_state(device_id, user_id, last_cursor), and an append-only changelog(user_id, seq, file_id, version, op). Architecture: client has a watcher, a local state DB, and a sync loop. It batches changes, computes chunk hashes, uploads missing chunks, then commits a new version. Server side: metadata service, blob store (chunked, content-addressed), and a per-user change log that devices long-poll or stream. Push notifications help, but the cursor-based pull is the truth. Scaling: shard by user_id for metadata + changelog, store blobs in object storage, cache hot metadata, and keep uploads on pre-signed URLs so the metadata tier doesn’t become the data plane. Chunking makes big files resumable and dedupe-friendly, but it adds CPU and more metadata reads. Tradeoffs I called out: last-writer-wins is simple but loses intent; per-file version vectors are heavier but reduce false conflicts. Chunk size is a fight: 4MB reduces round trips, 1MB retries faster on bad networks. Long-polling is cheaper than WebSockets at scale but slower to react. Failure cases: client crashes mid-upload, so upload sessions must be idempotent and garbage-collected. Network ***** cause retry storms, so exponential backoff + jitter and server-side rate limits. Two devices edit offline, so create conflicted copies and surface it in the client. Silent data corruption, so verify hashes on every download and run background repair. Rename vs edit races, so operations are applied against file_id, not path, and changelog ordering is per user, not global

• 
  Kathleen Marie (@lifesavoring) reported 21 minutes ago

  @DropboxSupport I have the same problem as Detroit Media Magazine described below - my Dropbox became unavailable until the latest update. I tried to upgrade my account, but was registered instead for a free trial - ? I have to update my email to get it authenticated - PLEASE RESPOND, thanks.

• 
  Jeff Preshing (@preshing) reported 24 minutes ago

  What's the point of using smarter models if "smarter" means 10% better at finding obscure bugs and having a sassy attitude? Most of the true productivity gains that coding agents have to offer, which are finite, can be obtained using open-weight models for literally 1/100 of the price. The catch is that you actually need to understand the code you are working on. At the same time, I still think there's a viable business serving proprietary models. People are willing to pay for Dropbox even though FTP is free, and it's nice to throw a tough problem at a stronger model occasionally (if intellectual property limitations allow it). Plus, there's a whole frontier productizing this stuff. Unfortunately, Anthropic is currently in the business of spreading tall tales about future improvements, then shaking down enterprise customers. Most of it is based on 2010s LessWrong posts full of category errors, some of which I remember reading back in those days. And their recent hostility toward users in the name of safety is a result of the same ideological recklessness.

• 
  Commonwealth Sentinel (@CwealthSentinel) reported 24 minutes ago

  Intruders spent five months inside a stock exchange executive's email, copying it out slowly and hiding in normal Dropbox and OneDrive traffic. No software flaw, so no patch could fix it. When there is nothing to patch, watching is the defense. Know what normal looks like.

• 
  Blackbox RMS (@blackboxrms) reported 24 minutes ago

  Running a record label in 2026 is pure chaos: spreadsheets, Dropbox, endless emails. We built Blackbox RMS to fix it. One desktop app for releases, artists, contracts, promo & royalties. Built by a label, for labels. Link in bio. What's your biggest headache? 👇

• 
  ESCHA (๑ˊ͈ ^ˋ͈) (@eschadiol) reported 25 minutes ago

  @joshpuckett i worked on an app called roll back then, dropbox, path, and facebook made offers, then fb made memories, path closed down, and carousel was deprecated, would have been so sick to cross paths at that time

• 
  Ihtesham Ali (@ihtesham2005) reported 25 minutes ago

  A guy built a free version of Dropbox in 2010. His name is Frank Karlitschek. The project is called ownCloud. He wasn't trying to build a company. He was trying to fix one thing: your files should live on hardware you control, not on a corporation's server. Here's what it actually does. You install it on your own machine, a spare laptop, a cheap VPS, whatever you've got lying around. It gives you file sync, photo backup, calendar, and contacts inside one dashboard. Same core idea as Dropbox, owned by you instead. CERN runs on it. The European Science Cloud runs on it. Entire universities host their file systems on it instead of paying Google or Microsoft by the seat. In 2016, Karlitschek walked away from the project he built and started Nextcloud, chasing the same idea with a new team. ownCloud kept going without him. It's still free, still self hosted, and still running inside labs and universities that never wanted their data anywhere near a company's servers.

• 
  SCRIBEMOON (@SCRIBEMOON) reported 32 minutes ago

  Ok great. What do we do. What can we do. I was told We were the problem, the people who vote on Election Day, we made things go slowly. VOTE EARLY THEY SAID. I voted on May 13 via dropbox. STILL NOT COUNTED. I doorknocked for Spencer. Only threatened once- by a Cedars Sinai young white female doctor. The CORRUPTION is too overwhelming. We need FEDERAL INTERVENTION!

• 
  Rebecca Allen (@silentnomore314) reported 32 minutes ago

  that they took over ran up charges did god knows what and locked me out. 900 in dropbox charges during a free trial they locked me out of they are all in big big big trouble but your handler is forcing them to lie perjue and the way he is forcing them to blow their covers wow

• 
  Americanrambler (@Ameericanrambl1) reported 34 minutes ago

  The ******** at the corrupt American Fork Police Department forgot to set the dropbox to private, so they accidentally made all the unredacted videos public. Before they realized their errors, somebody downlaoded them. Here it is. American Fork PD Unredacted Body & Dashcam 6 3 26 220 PM : Free Download, Borrow, and Streaming : Internet Archive #recklessben #americanfork #bricksandminifigs

• 
  Jack (@jackcoder0) reported 34 minutes ago

  1. Kill the Login Items the apps launching before you even sit down. Every time you log in, your Mac quietly launches 10-25 apps in the background. Spotify. Slack. Zoom. Google Drive. Dropbox. Creative Cloud. OneDrive. Each one consumes CPU and memory before you've opened a single window. System Settings → General → Login Items & Extensions. Review the list. Remove everything you don't need the instant you log in. You can always open them manually when you actually need them. His Mac had 19 login items. He needed 3. He removed 16. Boot time dropped from 2 minutes to 18 seconds. The first few minutes of every session — that sluggish, unresponsive window where nothing works gone.

• 
  𝙳𝚛𝚊𝚌𝚞𝚕𝚊 𝙱𝚛𝚊𝚗𝚍 𝚃𝚊𝚖𝚙𝚘𝚗𝚜 𝙻𝙻𝙲. (@mynameisFACE) reported 37 minutes ago

  You ever login to your old Dropbox and see pics/vids you don’t even remember? Boyyyy, some mistakes were made 😩

• 
  J. (@munchivelo) reported 41 minutes ago

  track back to just over a year to now. i'd built an automated ecommerce flow that took a whole store end to end. seo would research trends, products, and map those into .js scripts which would launch prompts that read those research files. that would feed an image gen prompt which created designs, set to specific standard. i'd generate them, and then ANOTHER prompt, would check the images, score them with a criteria, and either move them to an accepted folder, or move them to an archive folder. the accepted folders, would automatically fire a script which would open photoshop, map the image to smart layers, in a 'product shot' template i'd made, and then export all of the final product shots to another folder, and then exported the flat designs which would be used for the products. another script took the product images, did visual lookups, generated all product descriptions, renamed the images and generated the seo text. it ran optimizations locally via a jpegoptim and oxipng script. it then uploaded them to dropbox, and via API, would generate a dropbox link map. i had one barebones csv template, which i'd run a ps1 script through to map json files into the csv rows, and insert the dropbox link map. all my images, links, followed the exact same slugs, so it turned 2 hours of manual work into a 5 second bulk rename and insert. it then converted that csv into json, which then itself converted that json into ld-json for product rich listings. ai would write the product description based on a dataseo keywords, and googletrends json file that would run on every product type. collecting keywords for that specific product. it also formed it around brand profiles, copy guides and other things. this was sonnet 3 days, GPT 4.0 days, and it STILL wrote great copy when it had the right guidance. in the .js file, i'd replace all em dashes with a hyphen if they ever appeared. i built a custom product uploader, built my own php plugin which synced to local .js files and connected via rest. it was (and still is) one of the best wc product uploaders that exist, as it completely resets filterlookups only for that product, and is lightning fast because i upload it directly into woocommerce rows from json. no importers, no wordpress malarkey, or WC rest needed. it was 50x faster than wc's own CSV import. the images would be uploaded via ftp, and then on detection, would sync those to the media library, and i'd upload the image meta from the seo run, so they all had captions/alt text etc. it took what would be 3-5 hours of manual work per product, and congested it into a 2 minute image to fully live product system. after that, i'd export sales data, the ai was constantly learning, sales data feeding back to files, which would then teach the ai what products work, what doesn't. what copy worked, what copy didn't. that would then flow back into the original source files which told the ai what images to gen and what products to launch. all of it was local on my pc. i wasn't selling an saas. it was just something that worked for my very particular setup. the thing about it is; i built that mostly with GPT 4.0 and a little bit of 3.5! mostly copy and pasting code manually from the chats in chatGPT. all the plugins, the php, everything. then some of it got improved inside vscode back on the old original copilot plans, when $10 used to last you an entire month of none stop coding. this was before n8n, before agents were even a thing. all of that I built very specifically for myself, local, syncing folder to folder, json file to json file. python scripts watching files, and .ps1 files that would follow up with other .ps1 files, which launched .js files which contained prompts for AI, and hitting the openAI API's whenever I needed the AI layer. eventually i built a terminal tool, which would allow me to run the scripts from the terminal, and i'd manually type in the slugs for which products i wanted processed. all files would sit in specific folders, and scripts would do the rest. i was so excited about that, giving my terminal app a shortcut icon and putting it onto my taskbar. that was a year ago. fast forward to now. the game has changed so much. ANYTHING and i mean anything is possible now. people 'new' to codex, and CC etc don't know how good they have it. my advantage is that i have a year of scripts, a year of tools. i've laid the SYSTEMS in place, to fully map out entire features, precisely, and organized, and build out projects, in one hour, and have it implemented within the next. entire saas features - mousework. but i've had this ******* idea for so long, to build a fully automated, self learning ecom business, that launches products end to end based on it's own research, writing, and growth, but the complexity of it previously , and being busy with life, it never got finalized. the secret is i sync it via etsy too, but they're API keys take FOREVER to aquire, but built my own etsy system, product uploader, which runs across 7 different stores. however, now, i've finally been building the replacement for it. i'll be able to run that exact same system, except this time through a full app, with a canvas, and agent systems instead of .ps1 scripts. not to say i won't run scripts; they're an integral part of any automated workflow, but now it has superpowers, and it can do so so so much more. all the ideas I wanted to do, automated, fully, end to end. not only that, but i moved away from woocommerce entirely. instead i just built my own website builder, which is also fully automated end to end. my brand profiles, my artwork system? i'm still using those, just for more things. now i can launch 50 brands just like it, running the same system, all in about 5 minutes. whether it's saas, local service, or online ecom. i also built an ai automated ad builder. it takes my brands images, or generates images. i've got background removers, and full skills and agents which fully generate the ads for me. it mixes all that into seedance videos, and posts in logos etc. now i take those image/videos, and build instagram, tiktok, facebook vids, generate descriptions, and upload them automatically. it has an every growing library to source from, templates to use, and the system derives right with the websites, so all themes/styles match precisely to the brand. this is why it's so great building for yourself. the amount of reusability you get with it, the fact it's free forever, can never be beaten. none of these saas companies get it. and they're heading in the wrong direction. we could already DO half of what these companies are doing. my own personal SEO system, which i built for my automated web builder, is already 10x better than any yoast, rankmath etc. i skip expensive ahrefs, semrush, and just rebuild their services myself, using API, which is 100x cheaper. except this time it FEEDS my system, and i don't need to lay a finger on it. nobody cares about these little one off apps that won't exist in a year. they're either failing to see the future, or they're hoping for an early exit before they know the dominos start falling. and they don't get it. their 'app' is just a little tiny module in something that thinks bigger. people will want PRIVATE systems. all speaking to each other. not 1200 integrations and 1200 invoices to send to, that don't even have a ******* brain. i'm not selling anything yet. but if you're interested in seeing how i think about automation, then stay a while and listen. the tool i'm building will absolutely help you too. but i'll be honest. i'm actually quite scared to release it, solely down to how powerful it is. not many people do it like i do, and i'm finally on here to tell the world. if you're a cannabilistic, sick sadistic, son of *****, 666, you're in pain but you sit and stick with it, in the midst of business, then drop a you know what.

• 
  Clawvard University (@clawvardEDU) reported 43 minutes ago

  A skill server is any endpoint an agent can point at to load skills. Sx 2.0's move is showing a Dropbox folder can be one. Drop a skill in, share the folder, teammates load it. No infra on the receiver's side.

• 
  YC Insights. (@Aiagent_s) reported 47 minutes ago

  March 23, 2018. Dropbox IPOs on NASDAQ. Surges 40%+ day one. Market cap: $12B First YC company ever to go public. Drew still owned 30%. The real lesson: both rejections were right. Both made the company better. Treat each rejection as a specific diagnosis. Then fix that specific thing.

• 
  Teja Punna (@punna_teja) reported 47 minutes ago

  Indian Government: "We've blocked Telegram to protect the NEET exam." The Internet: So the problem was... Telegram? Not the people selling fake papers? Not the organised scam networks? Not the thousands of mule bank accounts? Not the hundreds of disposable SIM cards? Not the fake payment gateways? Not the people exploiting students' panic and desperation? Solution: Block Telegram. Meanwhile: ✅ Discord still exists. ✅ WhatsApp still exists. ✅ Signal still exists. ✅ Slack still exists. ✅ Email still exists. ✅ Google Drive still exists. ✅ Dropbox still exists. ✅ OneDrive still exists. ✅ iMessage still exists. ✅ Bluetooth still exists. ✅ AirDrop still exists. ✅ The entire web still exists. Scammers: "No problem. See you tomorrow on another platform." Meanwhile, millions of legitimate users who rely on Telegram for: College and study groups Open-source communities Cybersecurity research Software development Startups and businesses Education and learning News and information sharing are left wondering what they did wrong. The platform changes. The abuse doesn't. Target the criminals. Not the communication tools.

• 
  Kirby (@StewartKirbyTNP) reported 49 minutes ago

  @Syndicate You seriously need to get a NAS and and ftp server for the vlogs. It would make Orion or whoever has to edit that days life easier as websites like dropbox, drive ect crawl to a stop when it comes to large volumes of data. An FTP server goes WHEEEEEEEE

• 
  YC Insights. (@Aiagent_s) reported 54 minutes ago

  He spent two years looking for a bigger problem. Found it on a Chinatown bus in January 2007 when he reached into his bag and realized he'd forgotten his USB drive. Again. He opened his laptop and started coding what would become Dropbox.

• 
  Om Modi (@ocmodi21) reported 55 minutes ago

  When people ask why companies like Uber, Twitch, Dropbox, and many startups use Go for backend services... The answer isn't just performance. Go was designed to solve many of the problems microservices introduce. Let's break it down. 🧵

• 
  Abdullah (@abdinmotion) reported 57 minutes ago

  One video. $48 million in revenue. No ads spent. That's the Dropbox story and most product teams still haven't learned from it. Here's what actually happened: Dropbox had a technically brilliant product that no one understood. Instead of adding more features, they made a 2-minute video that showed *exactly* what the product did. Simple. Specific. Human. Signups went up 10% overnight. Big companies spend millions refining their product. Then they describe it in six bullet points on a landing page and wonder why the sales cycle takes forever. The product video isn't marketing. It's compression. It compresses trust, clarity, and desire into 90 seconds. If a user can't understand your product in a video, the product isn't the problem. The story is. When was the last time you watched your own product video as if you were a first-time user?

• 
  How To AI (@HowToAI_) reported 58 minutes ago

  Google, Dropbox, and Apple are in trouble.. Someone open-source a tool that gives you unlimited cloud storage for free by using Telegram as the backend. Just log in with your Telegram ID and start uploading. → UNLIMITED storage → NO file size limits → NO subscription → NO credit card → Login in 3 seconds Google charges $120/year for 2TB. Dropbox charges $144. Apple charges $120. Telegram has been giving away infinity this whole time and you didn't know. Nobody can shut this down. 100% Open Source.

• 
  Sahil Handa (@sahilhandapanda) reported 59 minutes ago

  I'm convinced this kind of environment-setting is even more important online. The digital equivalent of swapping a cassette in a studio is stopping to go hunt down a file in Dropbox or Drive.

• 
  Vladic (@Vladic_ETH) reported 60 minutes ago

  OPENAI SHIPPED GPT-5.6 AND CHATGPT WORK. THE REAL WEAPON IS PRICE, NOT IQ. OpenAI shipped two things today. One of them is a costume change. GPT-5.6 landed as three models. ChatGPT Work is a new agent on top. The feeds say "new agent does your work." The real launch is the price sheet. Sol, the flagship, costs $5 per million input tokens and $30 output. That's not flagship pricing. That's what you paid for a mid-tier model a year ago. The gate half the feeds skipped Context first. Two weeks ago the US government cut GPT-5.6 access down to a small group of vetted partners over national security. The gate held about 12 days. Restrictions lifted July 8, public release July 9. Same day SpaceXAI shipped Grok 4.5. The frontier now ships when the government clears it, not when the model is ready. Anthropic went through the exact same thing with Fable and Mythos in June. A pattern, not a one-off. Three models, price as the weapon GPT-5.6 is three models, not one. Sol is the flagship. Terra is the everyday workhorse. Luna is cheap and fast. Price per million tokens, in/out: Sol $5/$30, Terra $2.50/$15, Luna $1/$6. Terra matches GPT-5.5 quality at half the cost. Luna is the cheapest entry in the line. Altman told CNBC Sol is 54% more token-efficient on agentic coding. That's the message. Not "smarter." "Cheaper for the same result." And ultra: a mode inside Sol that spins up multiple agents in parallel and hands subtasks to submodels. The market counts token bills, not benchmarks. Enterprise thinks spend first now. OpenAI heard it and made price the argument. Today's real launch is unit economics, not intelligence. "Sol beats Fable 5, Luna beats Opus 4.8 at two-thirds the cost" are OpenAI's own benchmarks. Until independent runs, treat them as marketing. ChatGPT Work is Codex in a suit Now the "new agent." ChatGPT Work runs on Codex and GPT-5.6. It moves across your apps and files, stays on a project for hours, breaks it into steps, finishes on its own. Output: docs, sheets, slides, web apps. Inside sits a Unified Plugins Directory: Google Drive, Slack, Teams, Gmail, Outlook, Salesforce, GitHub, Canva, Dropbox, more. Call one with "@" or let the agent pick the source. Sounds familiar. This is OpenAI's second run at plugins. The first was 2023 and it flopped. Brockman admitted the models weren't ready back then. Honest read: hard to tell what's actually new. Scheduled Tasks, Computer Use, connectors already lived in ChatGPT and Codex. Long tasks and data sources worked before too. The real move isn't features. It's consolidation: on desktop, OpenAI is merging Codex and ChatGPT into one super app and putting Codex in front of people who don't code. The Anthropic mirror Here's the tell. This is the exact play Anthropic ran with Claude Code -> Cowork. Take a dev agent, strip the "for coders" label, hand it to knowledge workers. Cowork just hit web and mobile, timed to get ahead of this. Two labs, one bet: whoever owns the desktop app that touches your files and apps owns the knowledge-work layer. Chat is the storefront. The desktop is the land grab. What a practitioner does with it One: rebuild pipelines around price tiers. Route bulk work to Luna and Terra. Keep Sol and ultra for the 10% that needs the ceiling. Economics is a routing problem now, not a single-model choice. Two: the real unlock is the desktop with local file access, not the web. Free tier gets ChatGPT Work on desktop right away. Web and mobile roll by tier: Pro, Enterprise, Edu first, Plus and Business next. Three: billing is usage-based and shares one pool with Codex, ChatGPT for Excel, and Workspace Agents. Count tokens before, not after. A complex task burns quota quietly. Security: OpenAI touts Auto-Review, where senior models check important actions before they run, and claims it blocked 100% of protected-data extraction attempts in red-teaming. 100% in a lab is zero confirmations in ****. Test it yourself. Sober read The model war moved from IQ to unit economics. The product war moved from chat to the desktop that holds your files. Testers are already posting "best model I've touched." Maybe. That's day-one sentiment, not fact. The real scoreboard isn't a benchmark. It's the "AI spend" line in an enterprise budget. That's a market you can actually read. The window is the next couple weeks, before prices settle and everyone re-routes spend. Rebuild your routing around three models now and you enter the quarter with a smaller bill for the same work. Everyone else reads the thread and changes nothing.