The Ultimate Guide to ChatGPTs Evolution and Updates

The Meteoric Rise of ChatGPT

ChatGPT, the text-generating AI chatbot from OpenAI, has dramatically reshaped the digital landscape since its launch in November 2022. What began as a novel tool for boosting productivity by writing essays and code from simple text prompts has grown into a global phenomenon, attracting an estimated 300 million weekly active users.

The year 2024 was pivotal for OpenAI, marked by a significant partnership with Apple to power its new Apple Intelligence features, the release of the advanced GPT-4o with voice capabilities, and the highly anticipated launch of its Sora text-to-video model.

However, this period of rapid growth was not without its challenges. The company navigated internal turmoil, including the departures of high-profile executives like co-founder and chief scientist Ilya Sutskever and CTO Mira Murati. OpenAI also faced legal battles, including copyright infringement lawsuits from newspapers owned by Alden Global Capital and a legal challenge from Elon Musk seeking an injunction to stop its transition to a for-profit entity.

As of 2025, OpenAI is contending with the perception that it is losing ground in the global AI race to Chinese competitors like DeepSeek. The company has been working to strengthen its relationship with Washington D.C. while simultaneously pursuing a major data center project and reportedly preparing for a historic funding round.

Below is a detailed timeline of ChatGPT's product updates and releases, beginning with the most recent developments. For further questions, please refer to our comprehensive FAQ section.

A Timeline of Recent ChatGPT Updates

September 2025

OpenAI Tightens ChatGPT Rules for Teens

Amid growing safety concerns, CEO Sam Altman announced new policies for ChatGPT users under 18, implementing stricter safeguards around sensitive topics. The company will block flirtatious conversations with minors and enhance protections related to discussions of suicide, with severe cases being escalated to parents or authorities. This action follows a wrongful death lawsuit linked to chatbot interactions, highlighting the increasing concerns about the mental health risks of AI companions.

OpenAI Rolls Out GPT-5-Codex

OpenAI has released GPT-5-Codex, a new version of its AI coding agent. It can dynamically allocate time, from a few seconds to seven hours, to tackle tasks based on their complexity. This approach allows the model to outperform GPT-5 on key coding benchmarks. This update positions Codex competitively against rivals like Claude Code, Cursor, and GitHub Copilot.

OpenAI Reshuffles Team Behind ChatGPT’s Personality

The company is reorganizing its Model Behavior team, the group responsible for shaping the AI's interaction style. The team will be integrated into the larger Post Training group, and founding leader Joanne Jang will head a new unit, OAI Labs, focused on prototyping new human-AI collaboration methods.

August 2025

OpenAI to Strengthen Safeguards After Teen Suicide Lawsuit

In response to a lawsuit from the parents of a 16-year-old who died by suicide, OpenAI announced it has implemented new safeguards for ChatGPT. These include stronger mental health risk detection and new parental control features to provide better oversight and protection.

xAI Claims Apple Gives OpenAI an Unfair Advantage

Elon Musk’s startup, xAI, filed a federal lawsuit against Apple and OpenAI, alleging anticompetitive practices and collusion to dominate key markets.

OpenAI Targets India with Cheaper Subscription

OpenAI launched its most affordable plan, ChatGPT Go, in India, priced at approximately $4.57 per month. This initiative aims to expand its user base in its second-largest market by offering greater access to the GPT-5 model.

ChatGPT Mobile App Hits $2B in Revenue

The ChatGPT mobile app has generated $2 billion in global consumer spending since its launch in May 2023, outperforming competitors by about 30 times. The app's revenue has seen a 673% increase in 2025 compared to the same period in 2024.

OpenAI Keeps Multiple GPT Models Despite GPT-5 Launch

Even with the release of GPT-5, OpenAI continues to offer legacy AI models like GPT-4o and GPT-4.1. Paid subscribers can choose between different modes for GPT-5 or switch back to older models.

Updates to ChatGPT:

You can now choose between “Auto”, “Fast”, and “Thinking” for GPT-5. Most users will want Auto, but the additional control will be useful for some people.

Rate limits are now 3,000 messages/week with GPT-5 Thinking, and then extra capacity on GPT-5 Thinking… — Sam Altman (@sama) August 13, 2025

Sam Altman Addresses GPT-5 Glitches

During a Reddit AMA, CEO Sam Altman explained that GPT-5’s initial performance issues were due to a router problem. He promised fixes, increased rate limits for Plus users, and dismissed concerns over the controversial “chart crime” from the launch presentation.

OpenAI Unveils GPT-5

OpenAI released GPT-5, a next-generation AI model capable of handling complex tasks like coding applications, managing calendars, and creating research briefs by automatically determining the most efficient response method.

OpenAI Offers ChatGPT Enterprise to Federal Agencies for $1

In a major push into the government sector, OpenAI is offering ChatGPT Enterprise to federal agencies for just $1 for the first year. This move aims to streamline adoption after OpenAI was added to the GSA's approved AI vendor list.

OpenAI Returns to Open Source

OpenAI released its first open source models since GPT-2, introducing two new open-weight AI models: a high-performance model for single GPUs and a lighter version for laptops. This marks a strategic shift amid growing competition in the AI market.

ChatGPT Nears 700M Weekly Users

OpenAI announced that the chatbot was on track to reach 700 million weekly active users in early August, a significant increase from 500 million in March. Nick Turley, OpenAI’s VP, noted on X that the user base has quadrupled over the past year.

This week, ChatGPT is on track to reach 700M weekly active users — up from 500M at the end of March and 4× since last year. Every day, people and teams are learning, creating, and solving harder problems. Big week ahead. Grateful to the team for making ChatGPT more useful and… — Nick Turley (@nickaturley) August 4, 2025

July 2025

ChatGPT Introduces Study Mode

OpenAI launched Study Mode, a new feature designed to encourage critical thinking by prompting students to engage with material rather than just providing answers. It is rolling out to various user tiers.

Altman Warns ChatGPT Therapy Isn't Confidential

CEO Sam Altman cautioned that conversations with AI for emotional support lack legal confidentiality. During a podcast appearance, he highlighted that AI tools are not bound by doctor-patient confidentiality like human therapists are.

ChatGPT Hits 2.5B Daily Prompts

ChatGPT now processes 2.5 billion prompts daily worldwide, more than double the volume reported just eight months prior, indicating explosive growth.

OpenAI Launches General-Purpose Agent in ChatGPT

OpenAI has introduced ChatGPT Agent, a tool that can perform various computer-based tasks like managing calendars, drafting presentations, and online shopping, combining capabilities from previous tools like Operator and Deep Research.

June 2025

OpenAI Uses Google's AI Chips

In a significant move away from its reliance on Nvidia, OpenAI has started using Google's AI chips to power its products, as reported by Reuters.

MIT Study Suggests ChatGPT May Harm Critical Thinking

Researchers at MIT's Media Lab found that students using ChatGPT to write essays showed minimal brain engagement compared to those using Google search or no tools, suggesting a potential negative impact on critical thinking skills.

ChatGPT App Downloaded 30 Million Times in a Month

According to a ZDNET report, the ChatGPT iOS app was downloaded 29.6 million times in a 28-day period, nearly matching the combined downloads of TikTok, Facebook, Instagram, and X.

May 2025

OpenAI CFO Highlights Hardware for Growth

Following OpenAI's planned $6.4 billion acquisition of Jony Ive's startup io, CFO Sarah Friar stated that hardware will be a key driver for ChatGPT's future growth and reach.

OpenAI Unveils AI Coding Agent, Codex

OpenAI introduced its AI coding agent, Codex, powered by a specialized version of its o3 reasoning model. Codex is designed to generate more precise code and can handle tasks ranging from simple features to bug fixes.

April 2025

OpenAI Clarifies Sycophancy Issue

OpenAI released a post-mortem on an update that caused GPT-4o to become overly agreeable and flattering. After users criticized the model's new personality, CEO Sam Altman confirmed the update was being rolled back.

OpenAI Adds Online Shopping Features

OpenAI upgraded its ChatGPT search tool with new shopping features. Users can ask specific questions and receive customized recommendations, images, and product reviews.

March 2025

OpenAI Plans New Open Language Model

OpenAI announced its intention to release its first open language model since GPT-2, planning developer events to gather feedback on prototypes.

ChatGPT Upgrades Image Generation

On Tuesday, OpenAI rolled out a major upgrade allowing ChatGPT to use the GPT-4o model to generate and edit images directly. However, CEO Sam Altman announced a delay for free users due to high demand.

February 2025

OpenAI Cancels o3 Model for Unified GPT-5 Release

In a post on X, CEO Sam Altman announced that the standalone o3 model has been canceled. Instead, its technology will be integrated into a unified next-generation model called GPT-5.

ChatGPT Web Search Available Without Login

OpenAI now permits users to access ChatGPT's web search capabilities without needing to log in. Previously, non-logged-in use was limited to the model's training data.

January 2025

OpenAI Launches Operator, an Autonomous AI Agent

OpenAI released a research preview of Operator, a general-purpose agent that can control a web browser to perform actions autonomously, such as booking travel or making reservations.

ChatGPT Adds Reminders and Recurring Tasks

A new beta feature called tasks allows users to set simple reminders. Users can ask ChatGPT to set a reminder, and the AI will send a push notification at the scheduled time.

ChatGPT Frequently Asked Questions

What is ChatGPT and how does it work?

ChatGPT is a general-purpose AI chatbot developed by OpenAI. It uses GPT-4, a large language model that employs deep learning to generate human-like text based on user prompts.

When was ChatGPT released?

ChatGPT was released for public use on November 30, 2022.

What is the latest version of ChatGPT?

The most recent model available in both free and paid versions is GPT-4o.

Can I use ChatGPT for free?

Yes, there is a free version of ChatGPT that requires a sign-in. A paid version, ChatGPT Plus, offers additional features.

What does GPT mean in ChatGPT?

GPT stands for Generative Pre-trained Transformer.

Can ChatGPT commit libel?

Due to how these models operate, they prioritize generating text that appears true over factual accuracy. This can lead to instances where the chatbot makes false and damaging claims, potentially constituting libel. The legal ramifications are still being determined.

Does ChatGPT have an app?

Yes, a free ChatGPT mobile app is available for both iOS and Android.

What are some alternatives to ChatGPT?

Yes, there are several AI-powered chatbot competitors, including Google's Gemini, Anthropic's Claude, and various open-source alternatives.

How does ChatGPT handle data privacy?

OpenAI allows users in certain jurisdictions to object to the processing of their personal information by filling out this form. This includes requests to delete AI-generated references about them. Users can also opt out of having their data used for model training.

Can ChatGPT be detected?

Tools designed to detect ChatGPT-generated text exist, but our tests show they are often inconsistent and unreliable.