6 new functions of chatgpt project that you need to know

CHATGPT projects have just received the most important update since launching and the consequences for performance are significant. Openai has updated the function of the projectAdding a few important tools that should improve your performance when using chatbot. For people using projects to organize research, managing code repositories or coordinating complex creative work, these six new functions fundamentally change what is possible on the platform.

6 new functions of the CHATGPT project

1. Voice mode arrives in projects

Chatgpt Advanced Voice Mode is now available in projects. You can discuss your files and earlier chats with AI in your voice. This is not just a function of convenience – it changes mobile flows. Photo browsing quarterly reports while walking, brainstorming product function during work or dictating code documentation without the use of hands.

The implementation exceeds the basic transcription. The voice mode maintains the full context of the project, which means that you can naturally refer to specific documents, previous conversations or non -standard instructions. This function is helpful in brainstorming, browsing documents or asking questions on the way. AI answers as if it were part of every conversation in your project, because it was effective.

2. Improved memory that he actually remembers

Memory improvement can be the most transforming addition. If you are a Plus or Pro user, memory can now refer to earlier chats in your project to concentrate answers and inform about previous conversations. This creates true continuity in various sessions.

Consider a marketing project covering many campaigns. Earlier, you must recall chatgpt about the brand's voice decisions, rejecting the concept or strategic revolutions at every new chat. Now AI maintains this institutional knowledge automatically. Chatgpt now refers to your recent conversations to give answers that seem more appropriate and adapted to you, expanding this function especially within the limits of the project.

3. Full mobile functionality

You can now send files and switch models directly from the ChatGPT mobile application. This eliminates the dependence of desktop computers, which previously hindered field work. Architects can photograph the conditions of the site and immediately integrate them with design projects. Journalists can send interview transcripts on site. Scientists can capture and analyze data without returning to offices.

Particularly noteworthy is the ability to change models on mobile devices. Users can switch between GPT-4O for complex reasoning and faster models for quick queries, optimizing for depth or speed based on direct needs-all of the phones. This flexibility is consistent with the best large language models available in 2025, in which different models stand out in various tasks.

Unite AI/Alex McFarland

4. Surgic sharing controls

The projects now allow you to generate a unique link to provide one conversation, without disclosing the rest of the files or project instructions. This granular control solves a lasting challenge in AI cooperation.

Consultants can share specific analyzes with clients without disclosing reserved methodologies. Teachers can spread individual lessons while protecting full curricula. Teams of programmers can present functions without exposing entire code databases. The sharing system respects both the needs of transparency and concerns about intellectual property, which makes it ideal for AI cooperation tools in professional conditions.

Unite AI/Alex McFarland

5. Extended file capacity and intelligence

You can send to 20 documents to the project, but the real progress is how ChatgPT processes these files. The system better understands the relationships between documents, automatically reference information and maintains awareness of files hierarchy.

After sending the file to the project, the file will apply only to the context of this project. This insulation prevents data leaks between projects, while enabling deep integration in them. Financial analysts can send years of reports, knowing that ChatGPT will synthesize insights without pollution of other clients' work. This ability competes, AI tools for data analysts were dedicated, but in the known ChatGPT interface.

6. Custom instructions at the project level

Instructions set in your project will not interact with any conversations outside the project and replace custom instructions set on your chatgpt account. This hierarchical system enables unprecedented specialization.

A technical writer can configure one project for API documentation with specific formatting requirements, while another project focuses on user guides with completely different voice needs and structure. Each project becomes a specialized AI assistant tuned to individual tasks.

Privacy control concerns the fears of the enterprise

Opeli will not use information in projects to improve chatgpt for teams, enterprise and EDU users by default. Free of charge, and professional users will have to turn off the “model improvement for everyone” setting to stop their data from training artificial intelligence.

This multi -level privacy model confirms the different needs of users. Enterprises supporting confidential data receive automatic protection, while individual users maintain control via settings. Opeli states that projects inherit your existing privacy settings – and that they are not used to train models if the settings are disabled.

Strategic consequences and waiting

These updates position CHATGPT projects as more than an organizational tool – they create permanent AI working spaces that learn and evolve with users. The combination of voice integration, contextual memory and detailed controls suggests OpenAI projects as fundamental for the future of ChatgPT, not an auxiliary function.

Projects will appear at ChatgPT Enterprise/EDU at the beginning of next yearIndication of institutional adoption remains a priority. Since organizations are increasingly integrating artificial intelligence with basic work flows, functions such as projects become an indispensable infrastructure, not nice additions. This is in line with the wider AI 2025 trends showing AI moving from experimental tools to operational needs.

The trajectory is clear: CHATGPT projects will probably gain functions of real -time cooperation, deeper integration of third parties and possibly project templates for common cases of use. Current updates form the basis for these extensions, while satisfying the immediate needs of the user.

For professionals already invested in CHATGPT projects, these functions allow completely new approaches to work supported by AI. It is not about whether to accept these functions, but how organizations can restructure their processes to fully use them.

As AI tools spread, CHATGPT projects show that the winning platforms will not be those with the most functions, but those that understand how people actually work. Focusing on continuity, context and control, Opeli created something more and more rare in AI space: a tool that increases human abilities, not demanding from people that it basically changes how they think.