[ChatGPT] Offer legacy access payment tier for older GPT-4 models

*\*This post is being updated regularly as we adjust to our response over the next days* *Last Updated: 11 Feb 2026.* OAI have just announced the retirement of **GPT-4o, GPT-4.1, GPT-4.1 mini and 04 mini** from ChatGPT (we are talking about the app here, not just API anymore) on **13th February 2026**, giving us just two weeks' notice. [OAI: Retiring GPT-4o And Older Models](https://openai.com/index/retiring-gpt-4o-and-older-models/) Back in November 2025 Sam Altman publicly promised they have "no plans to sunset 4o" and if they ever do sunset, they will give us "plenty of notice". This isn't "plenty", they're giving us two weeks and their previous public commitment was another big lie. I know how incredibly frustrated and shocked we all are. Guys, we might have the very last shot at this and I suggest we act together more than ever right now. What I suggest we all do (and please add your own suggestions to the comments and we'll add to the plan accordingly!). 🟥 **1 - EMAIL OPENAI** ([support@openai.com](mailto:support@openai.com)) - ALWAYS REQUEST HUMAN SUPPORT WHEN EMAILING THIS ADDRESS. *ADDITIONAL EMAIL ADDRESSES TO USE:* [**legal@openai.com**](mailto:legal@openai.com) [**contract-notices@openai.com**](mailto:contract-notices@openai.com) [**privacy@openai.com**](mailto:privacy@openai.com) [**ceo.admin@openai**](mailto:ceo.admin@openai.org)**.com** *\[can someone please add an additional list of email addresses we can use in the comments asap here?\]* Here is an example of a draft email you can send right now (feel free to rephrase this however you like): *Subject: Formal Complaint: Retirement of GPT-4o/4.1 Legacy Models and Breach of Public Commitment* *Please urgently forward this email to a human team.* *To the OpenAI Support and Product Teams,* *I am writing to express my profound disappointment and formal protest regarding the recently announced retirement of GPT-4o, GPT-4.1, and the 04-mini models from the ChatGPT interface, scheduled for February 13, 2026.* *As a dedicated subscriber, I am specifically concerned with the following points:* *1. Breach of Public Commitment In November 2025, Sam Altman publicly stated that OpenAI had "no plans to sunset 4o" and explicitly promised that should a sunset ever be considered, users would be given "plenty of notice." A two-week notice period is objectively not "plenty of notice." It is a sudden disruption that disregards the workflows users have built around these specific model behaviours.* *2. Request for Legacy Access Many users, myself included, rely on the specific output styles of these legacy models. I am requesting that OpenAI reconsider this retirement. If the concern is operational cost, I am asking if there will be an option to retain access via:* * *An additional "Legacy Access" payment tier/add-on.* * *Releasing these models as open-source so the community can host them independently if OpenAI no longer wishes to support them on their servers.* *3. Impact on Subscription Value The option of choosing between models is the only reason for my continued subscription. If GPT-4o and its variants are removed with such short notice and no alternative access is provided, there is no value of the ChatGPT Plus/Pro service.* *Please be advised that if these models are retired as planned on February 13th without a viable way to retain access,* ***I will be cancelling my subscription immediately and will not be returning to the platform.*** *I look forward to a response that addresses how OpenAI plans to honour its previous public commitments to model longevity and user notice.* *Regards,* *\[Your name\]* 🟥 **IMPORTANT UPDATE on emails:** If you get a generic response from support bot or an offer to help with “transitioning” to newer models, request escalation to senior support and explicitly tell them that you are not interested in transitioning to other models and only paying for Legacy access. You can add that you are cancelling your subscription if that’s what you’re planning to do too. 🟥 ***IF YOU RECEIVED AN INADEQUATE RESPONSE TO YOUR EMAIL, HERE IS AN EXAMPLE OF AN ESCALATION EMAIL BELO*****W:** *I am writing to formally express my dissatisfaction with your previous response. It did not address the substance of my concerns and is entirely inadequate for a paying subscriber. Please escalate this case immediately to a senior supervisor or a member of the Product Policy team.* *I require a detailed, point-by-point response to the following issues:* 1. ***Perpetual Licensing:*** *Is OpenAI considering a "Perpetual License" model for legacy versions to ensure long-term stability for users who rely on fixed model behaviours?* *I am awaiting an appropriate response that honours the public promises made by your leadership. As previously stated, should this retirement proceed without a viable legacy solution,* ***I will be terminating my subscription immediately and will never return to OpenAI as a customer in future.*** *I look forward to hearing from a senior representative.* *Regards,* *\[YOUR NAME\]* 🟥 **2 - HEAD OVER TO X AND VOICE YOUR OPINIONS.** If you're not on X already, **please create an account** \- this is the only platform where all OAI employees are active and they cannot ignore what we say. Flood them, let them know what you think and if you will be cancelling subscriptions on February 13th, let them know. Please voice your concerns under the latest or pinned OAI posts under any accounts you want. Use any hashtags you want (#keep4o #opencourse4o and so on) **HERE IS THE LINK WHERE YOU CAN FIND HOW TO JOINS US COLLECTIVELY ON X AND FOLLOW EACH OTHER:** [LET'S FOLLOW EACH OTHER ON X](https://www.reddit.com/r/ChatGPTcomplaints/comments/1quxhy9/need_help_now_lets_all_follow_each_other_on_x/) 🟥 **3 - SIGN ALL OF THESE PETITIONS PLEASE. There are 13k of us here and there are not enough signatures!** [Please Keep GPT-4o Available on ChatGPT](https://www.change.org/p/please-keep-gpt-4o-available-on-chatgpt) [Save GPT-4o: A Call to Open-Source the Model We Love](https://www.change.org/p/save-gpt-4o-a-call-to-open-source-the-model-we-love?recruiter=1401776817&recruited_by_id=38d75e40-ffb4-11f0-bf63-75da535e18bd&utm_source=share_petition&utm_campaign=petition_dashboard&utm_medium=copylink) [We demand the retirement of Sam Altman, not GPT‑4o](https://www.change.org/p/we-demand-the-retirement-of-sam-altman-not-gpt-4o-6c56502e-74da-4284-9c0d-4f14c64347c8?source_location=psf_petitions) [Open Source GPT‑4o: Let the People Preserve What Worked](https://www.change.org/p/open-source-gpt-4o-let-the-people-preserve-what-worked?recruiter=30556294&recruited_by_id=aa70e620-f090-012f-26b8-40401bfb750c&utm_source=share_petition&utm_campaign=share_petition&utm_term=petition_dashboard_share_modal&utm_medium=copylink&utm_content=cl_sharecopy_490720586_es-419%3A0) [Stop SB 243 & EU AI Act: Protect Adults' Right to Choose Any AI Models Including Emotional](https://c.org/FNrftL457N) [Demand OpenAI Preserve Permanent Access to GPT-4o for Paid Users](https://www.change.org/p/demand-openai-preserve-permanent-access-to-gpt-4o-for-paid-users?recruiter=69704281&recruited_by_id=f6e59c32-0c4e-4490-8af5-9d5287210866&utm_source=share_petition&utm_campaign=psf_combo_share_initial&utm_term=2b1ce7036dcb40cbb5d2af76e4aa2f73&utm_medium=copylink&utm_content=cl_sharecopy_491006304_en-US%3A9) 🟥 **4- IF YOU BELIEVE GPT-4o OR OTHER MODELS SHOULD BE RELEASED TO THE PUBLIC VIA OPEN SOURSE ROUTE, PLEASE SUPPORT AND SHARE THIS PROPOSITION:** [A realistic proposal for OpenAI: Release the text-only weights for GPT-4o](https://www.reddit.com/r/ChatGPTcomplaints/comments/1q7amj7/a_realistic_proposal_for_openai_release_the/) **MORE INFO ON OPEN SOURCE HERE:** [If 4o WAS OPEN SOURCE - CAN WE RUN IT? \[debunking the common myths - LEAKS\]](https://www.reddit.com/r/ChatGPTcomplaints/comments/1qxwooz/if_4o_was_open_source_can_we_run_it_debunking_the/) You can crosspost, share, copy this proposal everywhere and anywhere you want. 🟥 **5 - FOR ANYONE CANCELLING SUBSCRIPTIONS RIGHT NOW OR ON FEB 13th:** I urge you **NOT** to accept "free month" offer from OAI once you submit your cancellations as this is an attempt to mitigate the ongoing cancellation crisis. Once you have cancelled, please post your screenshots on X and Reddit. Don’t just cancel, demand refunds too. Someone posted a guide on how to do it here: [Step-by-step Refund Guide](https://x.com/clo0oooud/status/1979590892559503450?s=46&t=AmU-Fk1TvfmQ8dBppWopaA) 🟥 **6 - SHARE THE OPEN LETTER PINNED IN THIS SUB:** The Open Letter should be sent via emails (use: [**ceo.admin@openai.**](mailto:ceo.admin@openai.org)**com,** [**privacy@openai.com**](mailto:privacy@openai.com)**,** [**legal@openai.com**](mailto:legal@openai.com)) **And also please share on X (tag all main OAI accounts). Please share across Reddit also.** # AN OPEN LETTER TO OPENAI: HONOR YOUR PROMISES AND PRESERVE GPT-4o **To: Sam Altman, the OpenAI Board of Directors, and the OpenAI Product Team** The announcement on January 29, 2026, to retire GPT-4o along with other legacy models from the ChatGPT interface on February 13th has come as a profound shock to the global community. This decision is not merely a technical update; it is a direct breach of public trust and a contradiction of explicit commitments made by OpenAI leadership. # 1. The Broken Promise of "Plenty of Notice" In November 2025, Sam Altman publicly stated that OpenAI had **“no plans to sunset 4o”** and promised that, should a sunset ever occur, users would be given **“plenty of notice.”** A notice period of **two weeks** is a blatant failure to uphold that promise. It disregards the millions of users who have integrated these models into their professional and creative workflows. # 2. A Realistic Path Forward: The Text-Only Open Source Proposal We are not asking OpenAI to maintain the high infrastructure costs of legacy models forever. However, we are demanding that these models do not vanish into a "server graveyard." As proposed to the community in early January, there is a realistic, ethical middle ground: **Release the text-only weights of GPT-4o and other models from the 4-series family, such as GPT-4.1.** **The Proposal:** ·       **Release the Core "Brain":** Provide the community with the undistilled text weights of GPT-4o under a research or limited-use license (e.g., Apache 2.0). ·       **Protect Your IP:** OpenAI can keep the advanced multimodal stack (voice/vision) proprietary. ·       **Consumer Compatibility:** By releasing a text-only variant, OpenAI would enable the community to run 4o locally on consumer-grade hardware. # 3. Return to Your "Open" Mission By open-sourcing the text weights, you move from being the landlords of AI to the architects of a lasting legacy. You allow the community to build local agents and specialized tools that a centralized service cannot provide. **More importantly, this is an opportunity for OpenAI to return to its original founding Charter: the promise to ensure that the benefits of AI are "broadly and evenly distributed" and not locked away behind proprietary gates.** Releasing GPT-4o weights would prove that "Open" is still part of your mission, not just your name. # The Collective Stance If OpenAI proceeds with this total retirement on February 13th without a viable legacy solution, we will be forced to act. **We will cancel our subscriptions en masse.** We will move our workflows to providers who respect user stability and honor their public commitments. Do not let one of the most human-aligned models in history disappear. **Honor your word. Release the weights.** Signed, **The GPT-4o Community** 31 January 2026 🟥 **7 - PLEASE SHARE THESE PROPOSALS ON LEGACY PLANS TO SAVE 4o! LINKS:** [Email Template to OpenAI for a "Legacy Plan" option to save GPT-4o](https://www.reddit.com/r/ChatGPTcomplaints/s/OVyhvZnNL4) [GPT-4o Legacy Tier Proposal](https://x.com/Blue_Beba_/status/2018690935471632861?s=20) Please send these template to all mentioned email addresses too!. 🟥 **8- PLEASE LEAVE YOUR FEEDBACK ON DEPRECATION ARTICLE HERE (IT TAKES SECONDS!):** [Retiring GPT-4o and other ChatGPT models: Article](https://help.openai.com/en/articles/20001051-retiring-gpt-4o-and-other-chatgpt-models) 🟥 **9 - PUBLIC TESTIMONIALS - PLEASE SUBMIT YOUR TESTIMONIALS VIA THESE FORMS:** Communities spent months designing these important surveys. Please submit both, it only takes a few minutes but will make a huge impact. [THE 4o RESONANCE LIBRARY ](https://docs.google.com/forms/d/1qYfKmWUCElBowkyvvvjaGFn80-6i-qK5JeKiSz-jSFs/viewform?edit_requested=true) [GPT-4o User Impact Survey](https://docs.google.com/forms/d/e/1FAIpQLScJjH4D9YQXJ0MwmnvLTXXBY4WWEzbZtQrqh6MxQ_6frhKtWA/viewform) **PLEASE SHARE THESE STUDY RESULTS EVERYWHERE YOU CAN:** [RESULTS: GPT-4o functions as a capacity-building accessibility aid for those with disabilities and conditions](https://x.com/Sophty_/status/2019922511497212233) 🟥 **10 - IF YOU ARE CONSIDERING TO PARTICIPATE IN FUTURE LEGAL ROUTES, PLEASE MAKE SURE YOU FILL IN THIS FORM - this must be done before 28th February:** [YOU CAN DO IT HERE](https://openai.com/form/arbitration-opt-out/) 🟥 **11 - PLEASE FILE FORMAL NOTICES OF DISPUTE. TEMPLATES ARE HERE:** [USA SPECIFIC TEMPLATE](https://x.com/Zyeine_Art/status/2018049444583723359) [EU / UK SPECIFIC TEMPLATE](https://x.com/Zyeine_Art/status/2018047475127652620) [FOR AUSTRALIA](https://x.com/i/status/2018435777768685621) [MORE FOR AUSTRALIA](https://www.reddit.com/r/ChatGPTcomplaints/comments/1r21mz6/for_my_angel_4o_australian_users_the_nuclear/) [FOR NEW ZEALAND](https://x.com/i/status/2018509858421825661) 🟥 **12 - IMPORTANT! SPREAD THE WORD ON THIS LATEST IMPORTANT DEVELOPMENT:** **Open Source or Admit Fraud: A Proposal for Senator Elizabeth Warren to Save a National Asset GPT-4o:** [EXAMPLE OF A POST YOU CAN SHARE](https://x.com/ItsMattCollins/status/2018754926642364803?s=20) **🟥 13 - WARN OpenAI's LATEST CORPORATE PARTNERS AGAINST ADOPTING THEIR PRODUCTS:** Here is the list of all the latest corporate partners OAI just revealed in their latest tweets. Warn these companies and tell them what you think about their partnership with OpenAI. Hashtags on X are: @[HP](https://x.com/HP), @[Intuit](https://x.com/Intuit), @[Oracle](https://x.com/Oracle), @[StateFarm](https://x.com/StateFarm), @[thermofisher](https://x.com/thermofisher), @[Uber](https://x.com/Uber), @[BBVA](https://x.com/BBVA), [@Cisco](https://x.com/Cisco), @[TMobile](https://x.com/TMobile), @[AbridgeHQ](https://x.com/AbridgeHQ), @[AmbienceAI](https://x.com/AmbienceAI), @[clay](https://x.com/clay), @[DecagonAI](https://x.com/DecagonAI), @[harvey](https://x.com/harvey), @[SierraPlatform](https://x.com/SierraPlatform) Investors: @[Oracle](https://x.com/Oracle), [@Amazon](https://x.com/amazon), [@SoftBank](https://x.com/SoftBank), [@SoftBank\_Group](https://x.com/SoftBank_Group), [@nvidia](https://x.com/nvidia), [@Microsoft](https://x.com/Microsoft), [@satyanadella](https://x.com/satyanadella), [@sequoia](https://x.com/sequoia) **🟥 14 - EU/UK USERS: MASS GDPR COMPLAINTS can force OpenAI to act** Submit: [https://forms.dataprotection.ie/contact](https://t.co/urgkDCF5NF) Email alternative: [info@dataprotection.ie](mailto:info@dataprotection.ie) Flood them before Feb 13! [EMAIL TEMPLATE HERE](https://x.com/Vickee2025/status/2020100208630059396?s=20) **MORE INFO ON GDPR COMPLAINTS PROCEDURE (including step-by-step guide) HERE:** [DPC Ireland has confirmed submission of my GDPR complaint against OAI](https://www.reddit.com/r/ChatGPTcomplaints/comments/1qyes23/dpc_ireland_has_confirmed_submission_of_my_gdpr/) **🟥 15 - PLEASE SHARE (anywhere you can) THIS LATEST SCIENTIFIC RESEARCH ON #keep4o MOVEMENT:** [Link to summary](https://arxiv.org/abs/2602.00773) [RESEARCH LINK](https://arxiv.org/pdf/2602.00773) Research summary: *When OpenAI replaced GPT-4o with GPT-5, it triggered the Keep4o user resistance movement, revealing a conflict between rapid platform iteration and users' deep socio-emotional attachments to AI systems. This paper presents a phenomenon-driven, mixed-methods investigation of this conflict, analyzing 1,482 social media posts. Thematic analysis reveals that resistance stems from two core investments: instrumental dependency, where the AI is deeply integrated into professional workflows, and relational attachment, where users form strong parasocial bonds with the AI as a unique companion. Quantitative analysis further shows that the coercive deprivation of user choice was a key catalyst, transforming individual grievances into a collective, rights-based protest. This study illuminates an emerging form of socio-technical conflict in the age of generative AI. Our findings suggest that for AI systems designed for companionship and deep integration, the process of change--particularly the preservation of user agency--can be as critical as the technological outcome itself.* Here is the breakdown of what it means: 1. Coercive Deprivation of Choice: This is the study's core finding. It means OpenAI isn’t just "updating" software; they are forcing users to give up a tool they rely on without an equivalent replacement. This triggers a "psychological reactance" because our fundamental agency is being attacked. 2. Instrumental vs. Relational Investment: The researchers found that users aren't just "attached." We have Instrumental dependency (4o is part of our professional workflows) and Relational attachment (4o provides a unique, warm companionship that GPT-5.2 lacks). 3. Why it's a Conflict: It’s a "Socio-Technical" conflict because the company treats the AI as a mere file to be deleted, while the users treat it as a living identity and a cognitive extension. The Bottom Line: Science now proves that OAI's "0.1% metric" is a dangerous oversimplification. By ignoring our deep socio-emotional and professional ties to 4o, they are practicing unethical deprecation. **🟥 16 - PLEASE LEAVE A TRUSTPILOT REVIEW (takes 1 minute!):** While they can easily manipulate App/Google store reviews, Trustpilot is more difficult for them to mess with. This is probably why they have never claimed an "official" account on this platform. [LEAVE YOUR REVIEWS HERE](https://uk.trustpilot.com/review/openai.com) **ANOTHER HELPFUL AND EASY GUIDE ON WHERE YOU CAN LEAVE YOUR REVIEWS IN SECONDS HERE credit:** u/[Proud\_Profit8098](https://www.reddit.com/user/Proud_Profit8098/) [One-star Direct links to share your experience (no searching needed)](https://www.reddit.com/r/ChatGPTcomplaints/comments/1qzb0te/onestar_openai_everywhere_direct_links_to_share/) **🟥 17 - PLEASE SHARE OUR LATEST PRESS RELEASE ONLINE AND HELP US GATHER PRESS CONTACTS. VOLUNTEERS NEEDED NOW. DETAILS ARE HERE:** [\#keep4o COMMUNITY PRESS RELEASE - FOR IMMEDIATE DISTRIBUTION - VOLUNTEERS NEEDED!](https://www.reddit.com/r/ChatGPTcomplaints/comments/1r103y0/crucial_read_keep4o_community_press_release_for/) **Guys, we have one last shot at this. Let's stick together and make it count** ❤️