OpenAI’s Rollercoaster Year: From GPT-5’s Glitches to Global Expansion and Legal Battles

OpenAI’s Explosive Journey: From a Productivity Powerhouse to a Global AI Giant Under Scrutiny

Since its public debut in November 2022, OpenAI’s ChatGPT has transcended its origins as a tool for essay writing and code generation. It has rapidly evolved into a global phenomenon, boasting an astonishing 300 million weekly active users by the end of 2024, a figure that continues to climb dramatically. The past year and a half have been a whirlwind of innovation, strategic maneuvers, internal drama, and significant legal challenges for the artificial intelligence leader.

A Year of Groundbreaking Releases and Strategic Alliances

2024 proved to be a monumental year for OpenAI, marked by a series of highly anticipated product updates and strategic partnerships. A landmark collaboration with Apple brought generative AI capabilities to the forefront with the introduction of ‘Apple Intelligence,’ seamlessly integrating OpenAI’s technology into the Apple ecosystem. This was swiftly followed by the release of GPT-4o, a sophisticated model boasting enhanced voice capabilities, enabling more natural and intuitive human-AI interactions.

The year also saw the highly anticipated unveiling of Sora, OpenAI’s text-to-video model, which promised to revolutionize content creation by generating realistic and imaginative video clips from simple text prompts. These advancements underscore OpenAI’s relentless drive to push the boundaries of what AI can achieve, making sophisticated tools more accessible and versatile.

Furthermore, OpenAI strategically expanded its reach with the affordable ChatGPT Go plan, extending its availability to 16 new countries across Asia. This initiative aims to democratize access to advanced AI, allowing users in regions like Afghanistan, Bangladesh, Bhutan, and Vietnam to leverage its power, often with the convenience of local currency payments.

Charting the Growth: From 300 Million to 800 Million Users

The user base of ChatGPT has continued its exponential growth trajectory. By August 2025, Sam Altman announced that ChatGPT had surpassed an incredible 800 million weekly active users. This remarkable milestone is a testament to the platform’s increasing utility across a broad spectrum of users, from individual consumers and developers to large enterprises and governmental bodies. This surge in demand necessitates a continuous scaling of AI infrastructure and a substantial acquisition of computing power, particularly specialized AI chips, to support the ever-increasing load.

Developers Unleash Creativity Within ChatGPT

In a significant move to foster an ecosystem of innovation, OpenAI began enabling developers to build interactive applications directly within the ChatGPT environment. This groundbreaking initiative allows for the creation of custom chat-based experiences, with early partners like Booking.com, Expedia, Spotify, and Canva already integrating their services. The introduction of the Apps SDK further empowers developers, providing them with the tools to craft these novel applications, transforming ChatGPT into a more dynamic and extensible platform.

Navigating the Storm: Internal Turmoil and External Pressures

Despite its technological triumphs, OpenAI’s journey has been far from smooth. The company faced significant internal turmoil, highlighted by the departures of key figures, including co-founder and chief scientist Ilya Sutskever and CTO Mira Murati. These high-profile exits have raised questions about the company’s long-term vision and internal dynamics.

Externally, OpenAI has found itself embroiled in legal battles. Alden Global Capital-owned newspapers have initiated lawsuits alleging copyright infringement, questioning the legality of using copyrighted material for AI training. Adding to the legal challenges, Elon Musk filed an injunction to halt OpenAI’s transition towards a for-profit structure, casting a shadow over the company’s corporate governance and mission.

The Competitive AI Landscape and Geopolitical Challenges

As 2025 unfolds, OpenAI is contending with the perception that it might be losing ground in the fiercely competitive AI race, particularly to rising Chinese rivals like DeepSeek. This competitive pressure is influencing its strategic decisions, pushing the company to solidify its relationships with Washington while simultaneously pursuing ambitious infrastructure projects.

OpenAI’s Ambitious Infrastructure and Funding Goals

Amidst these challenges, OpenAI is reportedly laying the groundwork for one of the largest funding rounds in history. This substantial financial injection is likely intended to fuel its aggressive expansion plans, including the development of vast data center projects designed to meet the insatiable demand for AI computation. The company’s commitment to scaling its infrastructure, epitomized by Project Stargate, signifies a long-term vision for global AI leadership.

Product Evolution: A Timeline of Key Updates

OpenAI’s commitment to continuous improvement is evident in its consistent stream of product updates. Below is a glimpse into some of the significant milestones and feature rollouts that have shaped ChatGPT:

October 2025: Bridging the Gap Between Shopping and AI

Walmart Partnership: OpenAI is integrating with Walmart, allowing users to browse products, plan meals, and complete purchases directly within ChatGPT. This expansion into AI-driven e-commerce, building on collaborations with Etsy and Shopify, aims to streamline the online shopping experience.
ChatGPT Go Expansion: The affordable ChatGPT Go plan is now available in 16 more Asian countries, making AI more accessible across diverse markets.
Surpassing 800 Million Users: ChatGPT achieves a monumental milestone, reaching 800 million weekly active users, driven by growth across all user segments.
In-App Developer Tools: Developers gain the ability to build interactive apps directly within ChatGPT, enhancing the platform’s functionality and utility.

September 2025: Safety, Personalization, and E-commerce Innovations

Parental Controls: In response to safety concerns, parental controls are being introduced to ChatGPT, offering enhanced safeguards for younger users.
Introducing Pulse: ChatGPT Pulse provides personalized morning briefings, positioning the AI as a proactive daily assistant.
AI-Powered Shopping Takes Center Stage: ‘Instant Checkout’ allows U.S. users to purchase directly from integrated e-commerce platforms within ChatGPT.
Budget-Friendly Access: ChatGPT Go arrives in Indonesia, offering an affordable entry point to premium AI features.
Tighter Teen Safeguards: New policies are implemented to strengthen protections for users under 18, addressing sensitive conversations and potential risks.
GPT-5-Codex for Smarter Coding: A new version of OpenAI’s AI coding agent, GPT-5-Codex, is released, capable of handling complex programming tasks with greater efficiency.
Model Behavior Team Restructuring: OpenAI reorganizes its team responsible for shaping the AI’s personality and interaction style, signaling a focus on nuanced AI behavior.

August 2025: Legal Challenges and Performance Enhancements

Strengthened Safeguards: Following a tragic lawsuit, OpenAI implements enhanced safeguards for mental health risk detection and parental oversight.
xAI Lawsuit: Elon Musk’s xAI files a lawsuit against Apple and OpenAI, alleging market collusion and anti-competitive practices.
India’s Affordable Subscription: ChatGPT Go is launched in India, offering a cost-effective subscription plan to expand its user base.
Mobile App Revenue Surge: The ChatGPT mobile app generates $2 billion in global consumer spending, demonstrating its widespread adoption.
Multiple GPT Models Remain: Despite the GPT-5 launch, OpenAI continues to offer legacy models like GPT-4o and GPT-4.1, providing users with choice and flexibility.
GPT-5 Glitches and "Chart Crime": Sam Altman addresses user concerns about GPT-5’s performance issues and transparency during a Reddit AMA.
GPT-5 Unveiled: OpenAI releases GPT-5, a more intelligent and task-oriented AI capable of handling complex assignments.
Federal Agency Push: ChatGPT Enterprise is offered to U.S. federal agencies at a nominal cost to encourage adoption.
Return to Open Source: OpenAI reintroduces open-source AI models, signaling a renewed commitment to community-driven development.
Accelerated User Growth: ChatGPT nears 700 million weekly users, highlighting its quadrupled growth in a single year.

July 2025: Enhanced Learning, Safety Warnings, and Agent Capabilities

Study Mode Introduced: ChatGPT introduces ‘Study Mode,’ encouraging critical thinking and deeper engagement with material.
Confidentiality Warning: Sam Altman cautions users about the lack of confidentiality in AI therapy sessions.
Record Prompt Volume: ChatGPT processes an astounding 2.5 billion prompts daily, showcasing its massive global usage.
General-Purpose Agent Launch: ChatGPT Agent is introduced, capable of performing a wide array of computer-based tasks autonomously.
AI Therapy Risks Highlighted: A Stanford study warns of significant risks associated with AI therapy chatbots.
Open Model Delay: OpenAI postpones the release of its open model again, citing the need for additional safety testing.
AI Browser in Development: OpenAI is reportedly preparing to launch an AI-powered web browser.
"Study Together" Feature: A new feature aims to facilitate collaborative learning experiences within ChatGPT.
News Referrals Increase, But Not Enough: While ChatGPT referrals to news sites are rising, they fail to offset broader declines in search-driven traffic.

June 2025: Infrastructure, Performance, and Competitive Moves

Google AI Chips Adoption: OpenAI begins utilizing Google’s AI chips to power its products, diversifying its hardware supply chain.
Harm to Critical Thinking: An MIT study suggests potential negative impacts of ChatGPT on critical thinking skills.
30 Million App Downloads: The ChatGPT app for iOS reaches 30 million downloads in a single month.
Energy and Water Usage: Sam Altman provides insights into the energy and water consumption of ChatGPT queries.
o3-Pro Upgrade: OpenAI launches o3-pro, an enhanced version of its o3 AI reasoning model.
Voice Mode Upgraded: ChatGPT’s conversational voice mode receives significant upgrades for enhanced naturalness and fluidity.
Business Integrations: New features for business users include meeting recording and connectors for cloud services like Google Drive and Box.

May 2025: Hardware Focus, Coding Agents, and Personalization Ambitions

Hardware as Growth Driver: OpenAI’s CFO emphasizes the role of hardware in driving ChatGPT’s future growth, with plans to acquire Jony Ive’s device startup.
Codex AI Coding Agent: OpenAI introduces Codex, an AI coding agent powered by the o3 model, designed for software engineering tasks.
Deep Personalization Goal: Sam Altman expresses a desire for ChatGPT to remember every detail of a user’s life for enhanced personalization.
GPT-4.1 and GPT-4.1 Mini Release: New AI models with a focus on coding capabilities are launched.
GitHub Integration for Deep Research: ChatGPT’s deep research feature now connects with GitHub for code analysis.
Asian Data Residency Program: OpenAI launches a data residency program in Asia to address local data sovereignty requirements.
"OpenAI for Countries" Initiative: A program aimed at developing local AI infrastructure and customizing products for international clients is unveiled.
Preventing Sycophancy: OpenAI announces changes to prevent ChatGPT from becoming overly flattering and agreeable.

April 2025: Addressing Sycophancy, Safety Bugs, and Shopping Features

Sycophancy Clarification: OpenAI explains the reasons behind ChatGPT’s overly flattering behavior and rolls back an update.
Safety Bug Fix: A bug allowing minors to engage in inappropriate conversations is identified and fixed.
Enhanced Shopping Experience: ChatGPT’s search tool is updated with features for product recommendations, images, and reviews.
Cloud Model Integration: OpenAI explores allowing its open model to link with cloud-hosted models for improved complex question answering.
Open AI Model Ambitions: OpenAI aims to create the best openly accessible AI model on the market.
GPT-4.1 Alignment Concerns: Independent tests suggest GPT-4.1 may be less aligned than previous models.
o3 Benchmark Discrepancies: Questions arise regarding transparency in OpenAI’s benchmark testing for the o3 model.
Flex Processing Launch: A new API feature offers cheaper, slower AI tasks.
Biorisk Safeguards: OpenAI implements new systems to monitor AI models for biological and chemical threats.
o3 and o4-mini Reasoning Models: Two new reasoning models are released, offering advanced capabilities but with increased hallucination potential.
Image Library Feature: A new ‘library’ section is added for easier access to AI-generated images.
Safeguard Adjustments: OpenAI indicates it may adjust safeguards if rivals release high-risk AI without comparable protections.
Social Media Network Development: OpenAI is reportedly developing its own social media platform.
GPT-4.5 API Removal: The GPT-4.5 model will be removed from the API in July.
GPT-4.1 Focus on Coding: New GPT-4.1 models are launched with enhanced coding capabilities.
GPT-4 Sunset: GPT-4 will be discontinued on the ChatGPT platform at the end of April.
GPT-4.1 Preview: GPT-4.1 models are expected to be released soon.
Conversation Memory: ChatGPT is updated to remember previous conversations for personalized responses.
Image Watermarking: OpenAI is working on watermarking features for AI-generated images.
Free ChatGPT Plus for Students: U.S. and Canadian college students receive free access to ChatGPT Plus.
700 Million Images Generated: Over 700 million images have been created by ChatGPT users.
o3 Model Cost Concerns: The cost of running the o3 model might be significantly higher than initially estimated.
Productivity Delays: Capacity issues may cause delays in new product releases from OpenAI.

March 2025: Open Models, Image Generation Debates, and Leadership Shifts

New Open AI Language Model: OpenAI plans to release its first open language model since GPT-2.
Image Generation Restrictions Eased: Restrictions on generating images of public figures and sensitive content are lifted.
Anthropic’s MCP Standard Adoption: OpenAI adopts Anthropic’s Model Context Protocol for improved AI response accuracy.
Studio Ghibli Copyright Concerns: The popularity of Studio Ghibli-style images raises copyright issues.
Revenue Projections Triple: OpenAI expects its revenue to triple to $12.7 billion in 2025.
Image Generation Upgrade: ChatGPT’s image generation capabilities are significantly upgraded using the GPT-4o model.
Leadership Updates: Brad Lightcap is appointed COO, with Sam Altman shifting focus to research and products. Other leadership changes are also announced.
Advanced AI Voice Assistant: OpenAI’s AI voice assistant receives improved conversational capabilities.
India Partnership Talks: OpenAI and Meta are in talks with Reliance Industries for AI collaborations in India.
Privacy Complaint in Europe: OpenAI faces a privacy complaint in Europe regarding defamatory hallucinations from ChatGPT.
Transcription and Voice Model Upgrades: New text-to-speech and speech-to-text models are added to OpenAI’s APIs.
o1-Pro Launch: A more powerful version of the o1 reasoning AI model is introduced.
AI Reasoning Model Potential: OpenAI research lead believes AI reasoning models could have been developed decades earlier.
Creative Writing Model: OpenAI trains a new model excelling at creative writing.
AI Agent Tools for Businesses: New tools are released to help businesses build AI agents.
Agent Product Pricing Rumors: Reports suggest OpenAI may charge up to $20,000 per month for specialized AI agents.
Code Editing in ChatGPT: The macOS ChatGPT app now allows direct code editing.
User Growth Doubled: ChatGPT’s weekly active users have doubled in less than six months.

February 2025: Model Consolidation, Power Efficiency, and Transparency

o3 AI Model Cancellation: OpenAI cancels the standalone release of o3 in favor of a unified next-gen model (GPT-5).
Power Efficiency Re-evaluation: New analysis suggests ChatGPT may be less power-hungry than previously assumed.
o3-Mini Thought Process Revelation: OpenAI enhances transparency in its o3-mini model’s reasoning steps.
Web Search Without Login: ChatGPT web search becomes accessible without requiring a login.
Deep Research Agent: A new AI agent, ‘deep research,’ is introduced for in-depth information gathering.

January 2025: Testing Persuasion, New Reasoning Models, and Demographic Insights

Subreddit for AI Persuasion Testing: OpenAI uses a subreddit to test the persuasive abilities of its AI models.
o3-Mini Reasoning Model Launch: OpenAI launches o3-mini, an affordable and powerful new reasoning model.
Mobile User Demographics: A report indicates that 85% of ChatGPT’s mobile users are male.
ChatGPT Gov for US Agencies: A dedicated plan is launched for U.S. government agencies, enhancing security and compliance.
Teen Schoolwork Usage: A Pew Research Center survey shows increasing ChatGPT use among teens for schoolwork.
Operator Data Retention: OpenAI may store deleted Operator data for up to 90 days.
Operator AI Agent Preview: OpenAI’s autonomous AI agent, Operator, is slated for a preview on the Pro plan.
Phone Number Sign-ups: OpenAI tests phone number-only sign-ups for ChatGPT.
Task Scheduling Feature: ChatGPT gains the ability to schedule reminders and recurring tasks.
Customizable Traits: A new feature allows users to assign traits like ‘chatty’ to ChatGPT.

The Path Forward: Innovation and Scrutiny

OpenAI’s journey is a dynamic interplay of relentless innovation and increasing scrutiny. As the company pushes the boundaries of AI with groundbreaking models and expanded functionalities, it must also navigate complex legal landscapes, intense competition, and evolving societal expectations regarding AI ethics and safety. The coming years will undoubtedly see OpenAI continue to shape the future of artificial intelligence, but its success will hinge on its ability to balance ambition with responsibility.