Musk’s OpenAI threat, GPT-5 teased, AI trust gap

It’s Thursday,

Elon Musk is willing to drop his $97.4 billion bid for OpenAI—if its board scraps plans to go for-profit—but CEO Sam Altman has already shut that down, according to a court filing.

1: OpenAI’s roadmap: New models coming soon

OpenAI CEO Sam Altman just revealed plans to launch GPT-4.5 and GPT-5 within “weeks” and “months,” signaling another big leap forward for ChatGPT. Although the company originally hyped a stand-alone “o3” model, Altman says it will now be integrated directly into GPT-5, rather than offered on its own.

So what does this mean for users? For one thing, you’ll see fewer baffling model choices inside ChatGPT. Altman admits the current interface feels clunky, and he wants it to “just work” for users without forcing them to pick from a half-dozen obscure model names. Once GPT-4.5 arrives—codenamed “Orion”—it’ll be the last model that doesn’t offer chain-of-thought reasoning. After that, GPT-5 will take over with a more unified, powerful approach to AI responses.

As always, there’s a lot of buzz around whether these updates will truly move the needle in areas like reasoning, context understanding, and content generation. But if nothing else, the next few months promise a sharper ChatGPT experience without the guesswork of choosing the “right” model.

Presented by Tabnine

How to make any LLM meet your IP liability standards

Navigating IP liability with AI models can be complex, but it doesn’t have to be. In this on-demand webinar, Tabnine breaks down how to ensure any LLM meets your company’s IP and compliance standards—without unnecessary risks.

Learn how to:

• Identify and mitigate IP risks in AI-generated code

• Implement practical safeguards for your team

• Ensure compliance without slowing down development

Watch on demand now

2: Trust in AI is much higher in China than in the U.S.

A new report from Edelman reveals a striking gap in global attitudes toward AI. In China, 72% of people say they trust AI—a sharp contrast to the 32% who do in the United States. India tops the charts at 77%, while Nigeria sits at 76%. By comparison, most Western nations trail well behind, with Canada, Germany, the Netherlands, the U.K., Australia, and Ireland all showing lower trust levels than the U.S.

The implications are big: Widespread trust in places like China and India could accelerate AI adoption there, while skepticism in the U.S. and parts of Europe might slow things down. Demographics also play a role. Younger people tend to trust AI more than older generations, and higher-income respondents are more open to the technology than those with lower incomes. Political affiliation in the U.S. matters, too—Democrats are likelier than Republicans or independents to trust AI.

This erosion of trust isn’t limited to AI. Tech companies in general have lost public confidence over the past decade. Concerns about job displacement and AI-driven misinformation loom large, suggesting that businesses have some explaining to do. Edelman’s advice? Show how AI will reshape jobs and society for the better. If leaders can demonstrate clear benefits and address people’s fears, they may help reverse the trend of declining trust in technology.

3: Meta’s AI may not be as good for ad growth as investors think

Meta’s shares have soared 24% this year on hopes that AI-driven tools will keep Facebook and Instagram ad revenue growing. CEO Mark Zuckerberg has doubled down on AI investments, promising advanced targeting and automated ad creation. That’s helped ad sales jump 21% in the fourth quarter—outpacing rivals like Google and Snapchat.

But advertisers aren’t all convinced. They say Meta’s costs have risen steadily without a clear bump in performance. AI-powered features that promise easy content creation and personalization are common across the industry, making it tough for Meta to stand out.

Long term, Meta’s dominant position in social media ads could be threatened if people turn to chatbots like ChatGPT or Google’s Gemini for product recommendations, prompting marketers to explore new platforms. For now, many will stick with Meta out of habit. But with skepticism growing, it’s unclear whether Meta’s AI push can truly sustain its ad surge.

4: US and UK refuse to sign AI summit declaration

At a Paris summit, roughly 60 nations—including China, India, and Germany—agreed on a non-binding pledge to ensure AI is “safe, secure, and trustworthy.” But the United States and the UK declined to join. US Vice President JD Vance argued against “overly precautionary” rules that could stifle innovation, asserting that “the most powerful AI systems are built in the US.”

This refusal hints at widening divisions over AI governance. Europe is investing heavily in data centers to cut its reliance on the US and China while also pushing for stricter safeguards. Meanwhile, critics say America’s new stance departs from the previous administration’s more cautious approach, fueling concerns that global consensus on AI regulation remains elusive.

Editors picks ☕️

Financial Times: Make AI safe again.

The Verge: Scarlett Johansson calls for deepfake ban after AI video goes viral.

Financial Times: Alibaba says it will be Apple’s AI partner in China.