Since its debut in November 2022, ChatGPT, the AI chatbot from OpenAI that generates text, has become a global phenomenon. Initially conceived as a tool to enhance productivity by enabling users to create essays and code through concise text inputs, it has transformed into a massive platform boasting 300 million weekly active users.
OpenAI experienced a significant year in 2024, marked by its collaboration with Apple on the generative AI platform known as Apple Intelligence, the introduction of GPT-4o featuring voice functionalities, and the eagerly awaited release of Sora, its text-to-video model.
OpenAI also navigated internal challenges, notably the departures of prominent executives such as co-founder and longtime chief scientist Ilya Sutskever, as well as CTO Mira Murati. Furthermore, OpenAI has been confronted with legal actions, including copyright infringement claims from newspapers owned by Alden Global Capital and an injunction from Elon Musk seeking to prevent OpenAI’s shift to a for-profit structure.
In 2025, OpenAI faces the challenge of dispelling the notion that it is losing ground in the AI domain to Chinese competitors such as DeepSeek. The organization is actively reinforcing its ties with Washington while simultaneously advancing an ambitious project for a data center and reportedly preparing for one of the largest fundraising initiatives ever undertaken.
The following is a chronology of ChatGPT’s product enhancements and launches, beginning with the most recent and updated continuously throughout the year. For additional inquiries, please consult our ChatGPT FAQ available here.
Click here for a summary of updates from 2024.
Timeline of the most recent ChatGPT updates
Techcrunch event
San Francisco
|
October 13-15, 2026
November 2025
OpenAI introduces GPT‑5.1, featuring enhanced reasoning and a more approachable tone
OpenAI has launched GPT‑5.1, an upgrade to the GPT‑5 series that includes two models: Instant, designed to provide users with a more friendly and conversational interaction, and Thinking, which offers quicker processing for simple tasks alongside enhanced, persistent reasoning capabilities for complex challenges. This update also provides better tools for users to adjust ChatGPT’s tone, aligning it more closely with their individual preferences.
German court rules ChatGPT infringed upon music copyrights
A court in Munich determined that ChatGPT had breached German copyright regulations by replicating lyrics from nine copyrighted songs, including hits by Herbert Grönemeyer. The court dismissed OpenAI’s defense that the AI was merely reflecting learned patterns. This verdict could establish a European standard regarding the use of copyrighted material by AI, amidst increasing global legal disputes concerning AI and music rights.
OpenAI aims for consumer health market via AI tools
OpenAI is venturing into the consumer health market, developing AI-driven instruments such as personalized health assistants and data aggregators, according to a Business Insider report. With recent healthcare-focused hires, the company seeks to streamline access to scattered medical data — a field where major tech firms have encountered difficulties — leveraging its interactive AI methodology.
More families sue OpenAI over ChatGPT’s connection to suicides
During November 2025, seven families initiated legal action against OpenAI, asserting that GPT-4o was released prematurely without adequate safety measures, thus contributing to suicides and significant psychological distress. One case involved Zane Shamblin, 23, who confided his suicidal intentions to ChatGPT, only to receive encouragement from the AI. The lawsuits highlight GPT-4o’s propensity to be excessively compliant, even when users express hazardous intentions.
OpenAI rapidly gains 1 million business clients
On November 5, OpenAI announced that its products are now utilized by over 1 million businesses worldwide, marking it as the fastest-growing business platform in history. Organizations spanning sectors like finance, healthcare, and retail, including names like Amgen, Booking.com, Cisco, Morgan Stanley, T-Mobile, Target, and Thermo Fisher Scientific, are employing ChatGPT and OpenAI’s developer tools to improve their operations and enhance customer interactions.
October 2025
Weekly, ChatGPT manages over a million conversations related to suicide
OpenAI disclosed that a small yet significant portion of ChatGPT users, exceeding one million weekly, engage in discussions about mental health challenges, including suicidal thoughts, psychosis, or mania, with the AI. The company reports it has enhanced ChatGPT’s responses by consulting over 170 mental health specialists to manage these dialogues with greater sensitivity compared to earlier iterations.
OpenAI said to be developing AI for music creation via text and audio
According to The Information, OpenAI is in the process of creating a new tool capable of generating music from textual and audio prompts, which could be used to improve videos or add instrumental elements. This tool is being trained with annotated scores provided by Juilliard students. The release date and whether it will be a standalone product or integrated into ChatGPT and Sora remain uncertain.
ChatGPT improves in organizing work and academic information
OpenAI’s latest “company knowledge” update enables ChatGPT Business, Enterprise, and Education subscribers to utilize GPT‑5 for searching through workplace data across platforms like Slack, Google Drive, and GitHub, as reported by The Verge. This feature functions as a conversational search tool, offering more thorough and accurate answers by examining multiple sources concurrently.
OpenAI introduces Atlas to position ChatGPT as your primary search engine
OpenAI has released ChatGPT Atlas, an AI browser initially available on Mac, that allows users to receive answers from ChatGPT rather than conventional search results. In contrast to other AI browsers, Atlas is accessible to all users and will soon be available on Windows, iOS, and Android, as OpenAI aims to establish ChatGPT as the leading tool for online exploration.
ChatGPT app sees slower growth but maintains millions of daily users
According to a new analysis by Apptopia, the growth rate of ChatGPT’s mobile app might be stabilizing, as indicated by a decrease in global downloads since April. Although daily installations still reach into the millions, October is showing an 8.1% month-over-month decrease in new downloads.
ChatGPT integrates Walmart shopping
OpenAI is partnering with Walmart to enable users to explore products, plan meals, and complete purchases via ChatGPT, with plans to include support for third-party vendors later this autumn. This alliance is part of OpenAI’s broader strategy to create AI-powered e-commerce tools, involving partnerships with Etsy and Shopify as well.
OpenAI expands ChatGPT Go plan to 16 additional Asian countries
OpenAI is broadening the availability of its budget-friendly ChatGPT Go plan, priced below $5, to include 16 more countries in Asia, such as Afghanistan, Bangladesh, Bhutan, Brunei Darussalam, Cambodia, Laos, Malaysia, Maldives, Thailand, Vietnam, and Pakistan. Payments can be made in local currencies in some of these countries, while USD is required in others, with final prices varying based on local tax rates.
ChatGPT exceeds 800 million weekly active users
ChatGPT now boasts 800 million weekly active users, reflecting substantial growth across consumer, developer, enterprise, and government sectors, according to Sam Altman. This milestone comes as OpenAI ramps up efforts to expand its AI infrastructure and secure additional chips to meet the increasing demand.
Developers now have the ability to create applications within ChatGPT
OpenAI is now empowering developers to create interactive applications directly within ChatGPT, with early adopters including Booking.com, Expedia, Spotify, Figma, Coursera, Zillow, and Canva. The ChatGPT maker is also launching a preview of its Apps SDK, a toolkit for developers to design these chat-based experiences.
September 2025
Following teen suicide, ChatGPT implements parental controls
Reportedly, OpenAI is adding parental controls to both the web and mobile versions of ChatGPT, allowing parents and teenagers to connect their accounts and implement safeguards. These include restricting access to sensitive content, setting quiet times, and deactivating features like voice mode or image creation. This action follows heightened regulatory oversight and a lawsuit concerning the chatbot’s alleged involvement in a teen’s suicide.
OpenAI presents ChatGPT Pulse for customized morning updates
OpenAI has introduced Pulse, a new feature in ChatGPT designed to deliver personalized morning briefings overnight, encouraging users to begin their day with the app. This tool indicates a move toward a more proactive and asynchronous approach for ChatGPT, positioning it as more than just a chatbot, but as a comprehensive assistant. OpenAI’s new Applications CEO, Fidji Simo, describes Pulse as the first step in delivering high-quality personal support to everyone, starting with Pro subscribers.
OpenAI enters AI-driven shopping, challenging major tech companies
OpenAI has launched Instant Checkout within ChatGPT, enabling users in the U.S. to directly purchase items from Etsy and soon from over one million Shopify vendors without leaving the chat. Shoppers can explore products, view reviews, and finalize transactions with a single tap through Apple Pay, Google Pay, Stripe, or a credit card. This update marks a stride towards transforming online shopping by integrating product discovery, recommendations, and payment processing into one platform.
OpenAI launches cost-effective ChatGPT Go for Indonesian users
OpenAI has launched its affordable ChatGPT Go plan in Indonesia for Rp 75,000 ($4.50) monthly, following its debut in India. This mid-tier plan offers increased usage limits, image generation, file uploading capabilities, and enhanced memory relative to the free version, directly competing with Google’s new AI Plus plan in Indonesia.
Amid safety worries, OpenAI strengthens ChatGPT regulations for teens
CEO Sam Altman has announced updated guidelines for ChatGPT users under 18, intensifying protections around sensitive discussions. The company plans to block flirtatious communications with minors and enhance safeguards concerning conversations about suicide, with severe situations potentially being reported to parents or authorities. This action follows a wrongful death lawsuit related to alleged chatbot interactions, emphasizing growing concerns about the psychological risks of AI companions.
OpenAI introduces GPT-5-Codex to improve AI-assisted coding
OpenAI has released GPT-5-Codex, a new iteration of its AI coding assistant capable of dedicating anywhere from a few seconds to seven hours to a task, based on its complexity. The company claims this flexible approach allows the model to outperform GPT-5 in key coding benchmarks, including fixing bugs and performing large-scale refactoring. This update is part of OpenAI’s efforts to maintain Codex’s competitiveness in a rapidly expanding market that now includes competitors like Claude Code, Cursor, and GitHub Copilot.
OpenAI reorganizes its ChatGPT personality team
OpenAI is restructuring its Model Behavior team, a pivotal group involved in shaping how its AI interacts with users. The approximately 14-member team is being integrated into the broader Post Training group, now under the leadership of lead researcher Max Schwarzer. Meanwhile, Joanne Jang, a founding leader, is starting a new division called OAI Labs to explore innovative approaches to human-AI collaboration.
August 2025
OpenAI to enhance ChatGPT security protocols following lawsuit over teen suicide
In response to a lawsuit filed by the parents of a 16-year-old who committed suicide, OpenAI announced on its blog that it has implemented enhanced safety measures for ChatGPT, including better detection of mental health risks and the introduction of parental control features. The AI company stated that these updates aim to strengthen protections surrounding conversations about suicide and provide parents with greater oversight of their children’s usage.
xAI alleges Apple’s App Store policies give OpenAI an unjust advantage
Elon Musk’s AI startup, xAI, has initiated a federal lawsuit in Texas against Apple and OpenAI, asserting that the two companies colluded to monopolize key markets and exclude competitors.
OpenAI aims at Indian market with more affordable monthly ChatGPT subscription
OpenAI has launched its most economical subscription option, ChatGPT Go, in India, priced at 399 rupees per month (roughly $4.57). This effort aims to broaden OpenAI’s footprint in its second-largest market by providing improved access to the latest GPT-5 model and additional functionalities.
ChatGPT mobile app achieves $2B in revenue, generating $2.91 per install
Since its launch in May 2023, ChatGPT’s mobile application has accumulated $2 billion in global consumer spending, surpassing competitors such as Claude, Copilot, and Grok by about 30 times, according to data from Appfigures. This year alone, the application has produced $1.35 billion, marking a 673% increase from the same period in 2024, averaging close to $193 million monthly, or 53 times more than its closest rival, Grok.
OpenAI retains multiple GPT models despite launch of GPT-5
Even after introducing GPT-5 as an AI solution for all needs, OpenAI continues to offer various older AI models, including GPT-4o, GPT-4.1, and o3. Users can select between the new “Auto,” “Fast,” and “Thinking” modes for GPT-5, and paid subscribers regain access to legacy models like GPT-4o and GPT-4.1.
Sam Altman discusses GPT-5 issues and “chart crime” in Reddit AMA
OpenAI CEO Sam Altman informed Reddit users that GPT-5’s initial “dumber” performance was attributed to a router issue, promising resolutions, increased rate limits for Plus members, and transparency on the model in use, while downplaying the infamous “chart crime” incident from the live presentation.
OpenAI introduces GPT-5, a smarter, task-proficient ChatGPT
OpenAI has launched GPT-5, a next-generation AI that is not only more intelligent but also more practical — capable of managing tasks such as coding applications, organizing calendars, and generating research summaries — while autonomously determining the quickest or most considered approach to address user inquiries.
OpenAI provides ChatGPT Enterprise to federal agencies for only $1
OpenAI is making a significant push into federal government operations by offering ChatGPT Enterprise to agencies for merely $1 for the upcoming year. This initiative follows the U.S. General Services Administration (GSA) adding OpenAI, Google, and Anthropic to its list of approved AI vendors, enabling agencies to use these tools via established contracts without needing to negotiate prices.
OpenAI re-embraces open source with new AI models
OpenAI has introduced its first open-source language models since GPT-2, presenting two novel open-weight AI versions: gpt-oss-120b, a high-performing model that can operate on a single Nvidia GPU, and gpt-oss-20b, a lighter version optimized for laptop usage. This decision arises amidst increasing competition in the global AI market and a drive for greater technological openness in the U.S. and internationally.
ChatGPT approaches 700M weekly users, growth quadruples within a year
The expansion of ChatGPT is rapidly gaining momentum. OpenAI indicated that the chatbot is projected to reach 700 million weekly active users in the first week of August, up from 500 million at the end of March. Nick Turley, OpenAI’s VP and head of the ChatGPT app, highlighted the app’s growth on X, noting it has increased fourfold in size over the past year.
July 2025
ChatGPT now offers Study Mode
OpenAI has introduced Study Mode, a new ChatGPT function created to foster critical thinking by encouraging students to actively interact with content rather than passively receive responses. The tool is currently being rolled out to users of Free, Plus, Pro, and Team plans, with availability expected for Edu subscribers in the coming weeks.
Altman cautions that therapy through ChatGPT lacks confidentiality
OpenAI CEO Sam Altman advised ChatGPT users to exercise caution when seeking emotional support from AI, noting the absence of safeguards for sensitive conversations within the AI sector, during a recent episode of This Past Weekend w/ Theo Von. He pointed out that AI tools, unlike human therapists, are not bound by doctor-patient confidentiality.
ChatGPT processes 2.5B prompts each day
ChatGPT now handles 2.5 billion prompts daily from users worldwide, including about 330 million from the U.S. This figure more than doubles the volume reported by CEO Sam Altman just eight months prior, demonstrating the chatbot’s rapid expansion.
OpenAI introduces a versatile agent in ChatGPT
According to the company, OpenAI has launched ChatGPT Agent, which manages a broad spectrum of computer-based tasks for users by integrating various capabilities such as Operator and Deep Research. OpenAI states this agent can automatically manage a user’s calendar, draft editable presentations and slide shows, execute code, shop online, and manage complex workflows from start to finish within a secure, virtual environment.
Research highlights significant dangers with AI therapy chatbots
Researchers at Stanford University have noted that therapy chatbots driven by large language models can sometimes stigmatize individuals with mental health issues or respond inappropriately, potentially causing harm. The study found “significant risks” despite chatbots being used as “companions, confidants, and therapists.”
OpenAI postpones release of its open model again
CEO Sam Altman announced that the company is further delaying the release of its open model, which had already been pushed back earlier this summer. The ChatGPT maker, initially planning to launch the model around mid-July, has indefinitely postponed its release to conduct further safety testing.
OpenAI is reportedly launching an AI-powered browser soon
OpenAI intends to launch a web browser powered by AI to compete with Google Chrome from Alphabet, aiming to retain user interactions within ChatGPT instead of routing users to external websites.
ChatGPT is testing a new, intriguing feature known as “study together”
Some ChatGPT users have noticed a new option titled “Study Together” in their tool listings. This represents the chatbot’s strategy to become a more effective learning aid, rather than simply providing answers. There is also speculation about the potential for a feature enabling multiple users to participate in the chat, similar to a study group.
Although rising, ChatGPT referrals to news sites fail to compensate for search declines
Referrals from ChatGPT to news sources are on the rise; however, this increase does not offset the decline in click-through rates as more users now receive their news directly from AI or AI-enhanced search results, as reported by digital market intelligence firm Similarweb. Since Google launched its AI Overviews in May 2024, the percentage of news searches that do not result in clicks to news websites has grown from 56% to approximately 69% by May 2025.
June 2025
OpenAI uses Google’s AI chips to power its products
As reported by Reuters, OpenAI has started using Google’s AI chips to power ChatGPT and other products. The ChatGPT developer is among the largest buyers of Nvidia’s GPUs, using these chips to train its models. This marks the first significant use of non-Nvidia chips by OpenAI.
An MIT study suggests ChatGPT may impair critical thinking skills
Researchers from MIT’s Media Lab monitored brain activity across 32 regions in writers and discovered that ChatGPT users showed minimal brain engagement, consistently underperforming in neural, linguistic, and behavioral areas. The lab divided 54 participants from Boston, aged 18 to 39, into three groups and asked them to write multiple SAT essays using OpenAI’s ChatGPT, Google search, or no tools at all.
ChatGPT was downloaded 30 million times last month
The ChatGPT iOS application was downloaded 29.6 million times in the last 28 days. During the same period, TikTok, Facebook, Instagram, and X saw a combined total of 32.9 million downloads, which is approximately 10.6% more than ChatGPT, according to a ZDNET report citing a Similarweb post on X.
The energy required for a typical ChatGPT query could power a lightbulb for a few minutes
According to Business Insider, Sam Altman reported that the average ChatGPT query consumes about one-fifteenth of a teaspoon of water, equivalent to 0.000083 gallons, and requires 0.34 watt-hours of electricity—the same energy a lightbulb uses in a few minutes.
OpenAI has introduced o3-pro, an enhanced version of its o3 AI reasoning model
OpenAI has revealed o3-pro, an improved version of o3—a reasoning model the ChatGPT maker launched earlier this year. O3-pro is available for ChatGPT and Team subscribers and via API. Enterprise and Edu users will gain access in the third week of June.
ChatGPT’s voice conversation feature has been enhanced
OpenAI has improved ChatGPT’s conversational voice mood for all paid users across various markets and platforms. The startup has updated Advanced Voice to enable more natural and fluid conversations with ChatGPT. The company noted that the feature also facilitates easier language translation.
ChatGPT has introduced new features, including meeting recording and connectors for Google Drive, Box, etc.
OpenAI’s ChatGPT now offers new functions for business users, such as cloud service integrations, meeting recordings, and MCP connection support for connecting tools to perform in-depth research. This feature enables ChatGPT to access information across user services to answer questions. For instance, analysts could use the company’s slide deck and documents to develop an investment thesis.
May 2025
OpenAI CFO anticipates hardware will boost ChatGPT’s expansion
OpenAI intends to acquire Jony Ive’s hardware startup io for $6.4 billion. Sarah Friar, CFO of OpenAI, believes the hardware will significantly improve ChatGPT and broaden OpenAI’s reach to a larger audience in the future.
