This week’s stories come from France, Taiwan and USA, two of them are about AI and the third one is about chips

1-Nearly 500,000 California students to access to ChatGPT

The California State University (CSU) system is rolling out OpenAI’s ChatGPT Edu—a specialized version of ChatGPT designed for educational institutions—to over 460,000 students and 63,000 faculty and staff across its 23 campuses.

I read this story at Axios and ChatGPT is already reshaping higher education by enhancing students’ access to expert knowledge and improving efficiency. However, concerns remain about the long-term impact of generative AI on learning.

CSU faculty can leverage ChatGPT to design curricula and create custom course-specific GPTs, while students can use it for personalized tutoring, study materials, and research, according to the story.

Administrative staff can develop custom GPT models to assist students with enrollment and program-related inquiries. CSU will also connect students with AI-focused apprenticeship opportunities. After initial restrictions on AI tools, universities are now embracing AI education to enhance students’ career readiness, the story noted. (By the way I have a story here about AI).

Despite this AI integration, CSU faces proposed budget cuts that could reduce state funding by $375 million, the story put it. Let me add that OpenAI did not disclose the financial details of the agreement but noted that ChatGPT Edu provides security and controls similar to its Enterprise edition while remaining affordable for educational institutions.

Nearly 500,000 California students to access to ChatGPT

2-TSMC fast-tracks 1nm production, plans to establish “Giga Fabs” in Taiwan

TSMC is gearing up to reshape the market by preparing 1nm production lines in a cutting-edge facility, according to Taiwan Economic Daily. Reports indicate that TSMC plans to mass-produce its 1nm chips at a dedicated site in Tainan, Taiwan. The facility, named “Fab 25,” will feature six production lines manufacturing 12-inch wafers. In addition to 1nm, TSMC is also constructing new fabs for its upcoming 2nm and 1.4nm processes in Tainan, a region that benefits the company through government incentives and its growing status as a semiconductor hub, often compared to Silicon Valley.

I read this story at Wcctech and TSMC’s roadmap for 1nm was outlined at the IEDM conference, where the company projected its launch by 2030. The firm expressed confidence in achieving a groundbreaking “trillion transistors” milestone using advanced 3D-stacked chipsets. Notably, TSMC has adjusted its naming convention after 2nm, with 1.4nm and 1nm labeled as A14 and A10, resembling Intel Foundry’s approach, the story noted.

However, achieving this milestone presents challenges, particularly in yield rates and supply chain stability, issues that have intensified as semiconductor processes continue to shrink. Initial estimates suggested that TSMC’s 1nm plans would require an investment surpassing approximately $32 billion, a figure that has likely increased. With a projected timeline of five years, TSMC still has time to navigate these complexities and refine its approach, the story put it.

TSMC fast-tracks 1nm production, plans to establish “Giga Fabs” in Taiwan

3-France’s response to US’s Stargate Project: $112B AI Investment Initiative

French President Emmanuel Macron revealed plans for a substantial €109 billion ($112 billion) private investment in the country’s artificial intelligence (AI) sector last week. Macron emphasized that France and Europe are accelerating their AI development efforts. He compared the investment package to the U.S.’s $500 billion Stargate initiative, noting that given France’s smaller population (68 million), the funding is proportionate in scale.

I read this story at TechCrunch and it has tracked various investment commitments from international and domestic sources, amounting to as much as €83 billion ($85 billion) so far. These include contributions from the UAE (between €30B-€50B), Canadian firm Brookfield (€20B), French state investment bank Bpifrance (€10B), and telecom giant Iliad (€3B).

Arthur Mensch, CEO and co-founder at Mistral, announced his company’s intention to invest billions in a new AI computing hub in Essonne. Mistral, one of Europe’s leading AI firms developing foundational models, aims to compete with industry leaders such as OpenAI, Meta, and Alibaba. Mensch stated that this investment will allow for training more advanced AI systems in a shorter time frame.

Let me add that France’s commitment to AI is further strengthened by its energy strategy. With the majority of its electricity generated by nuclear power—producing more than it consumes—the government has pledged to allocate one gigawatt of nuclear power to AI training by 2026.

In 2024, France exported 90TWh of electricity to neighboring nations, and it now plans to leverage its energy surplus to attract foreign investors.

France’s response to US’s Stargate Project: $112B AI Investment Initiative (Photo: AFP)