ChatGPT users are getting a major upgrade this week with three new AI models

• OpenAI will launch GPT-5.6 after the White House cleared it following weeks of security reviews.

• The AI model was delayed over concerns that advanced systems could be exploited for cyberattacks and national security threats.

• GPT-5.6 will launch in three versions: Sol, Terra and Luna, each targeting different user needs.

• OpenAI says the new model includes stronger safety protections while offering improved performance in cybersecurity, biology and autonomous AI tasks.

If you've ever complained that ChatGPT gave you the wrong answer, forgot your previous prompt or simply wasn't smart enough, OpenAI believes its next update could change that.

The company is set to launch its new GPT-5.6 AI model series on Thursday after a delay caused by security concerns from the United States government.

According to Axios, the White House asked OpenAI to hold off on releasing the model in June while officials carried out technical testing and reviewed possible national security risks. After weeks of discussions with OpenAI executives, the government has now cleared the release.

OpenAI says GPT-5.6 is its most powerful AI model yet, with stronger abilities in cybersecurity, biology and autonomous AI tasks, also known as agentic AI. These are AI systems that can complete more complex tasks with less human guidance.

The delay wasn't because the model wasn't ready.

US officials have become increasingly worried that advanced AI systems could be misused by cybercriminals or foreign governments. One major concern is that highly capable AI can identify weaknesses in software code that hackers or intelligence agencies could exploit.

The same concerns recently affected another AI company, Anthropic. In June, the White House ordered the company to pause the use of its advanced Fable and Mythos AI models over national security concerns before easing the restrictions last week.

OpenAI agreed to delay GPT-5.6 while the review was carried out.

However, the company made it clear that it doesn't want this kind of approval process to become standard.

In a statement, OpenAI said it doesn't believe "this kind of government access process should become the long-term default" because it "keeps the best tools" away from users, businesses and other people who need them.

READ ALSO: Why AI agents are the next big leap after smartphones

Although US President Donald Trump has generally supported fewer regulations for artificial intelligence, his administration has also introduced measures to address security risks linked to increasingly powerful AI systems.

In June, Trump signed an executive order that creates a voluntary process allowing AI companies to share new models with the US government for up to 30 days before releasing them publicly.

OpenAI said it is now working with government officials "to develop the cyber Executive Order framework and a repeatable process for future model releases."

READ ALSO: AI won’t take your job. But someone using AI might

OpenAI says the GPT-5.6 family will launch with three different versions.

• Sol will be the flagship model and is described as the company's "strongest model yet."

• Terra is designed for everyday use and balanced performance.

• Luna will be the fastest and most affordable version, focusing on speed and efficiency.

The company also says GPT-5.6 includes a "layered safeguard stack," which combines multiple security systems designed to reduce the risk of the AI being used for cyberattacks or other harmful activities.

For millions of ChatGPT users around the world, including many Nigerians who rely on the chatbot for work, school, coding, writing and business, the big question now is simple: will GPT-5.6 finally be the AI upgrade everyone has been waiting for?