AI: 22 Countries Seek a Global Body Amid Hype and Flaws

Amid calls for regulation, stock market records, and successful applications, announcements surrounding AI and technology are piling up. Many of these come with energy constraints, vulnerabilities, or uncertainties. Timnit Gebru and Emily M. Bender urge caution, citing incentives to exaggerate AI capabilities.
Maritime and Biodiversity: Quantified Paths for Climate
Maritime transport accounts for about 3% of annual greenhouse gas emissions. Among the proposed solutions, wind energy is making a comeback: in the Marshall Islands, a cargo sailing ship inspired by traditional vessels made its maiden voyage last year and could, according to its designers, reduce emissions by up to 80% compared to a fuel-powered cargo ship. In the UK, projects for tree vaccines and genetically modified butterflies have received biotech funding, aiming to enhance ecosystem adaptation to climate change.
Rules of the Game: 22 Countries Seek Global AI Oversight
Twenty-two nations are calling for the establishment of an international body to oversee AI. They are advocating for the implementation of pre-deployment testing procedures and the adoption of common safety standards. The United States and China are not included among the countries that signed this declaration. Meanwhile, OpenAI has proposed its own safety standards and wishes for the United States to lead an international effort on this issue.
Infrastructure Under Strain: Data Centers and Space Launchers
In the United States, two states are tightening regulations on data centers: Texas has paused the issuance of new permits pending an audit of its electrical grid, while California has instituted new regulations regarding water and energy consumption. In the space sector, a shortage of launchers is making access to orbit more difficult for satellites, as SpaceX slows down its Falcon 9 launches and other players struggle to keep up.
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisissez votre rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Market and Uses: Record Valuation, Leading App, and Highlighted Flaws
Driven by demand for AI chips, AMD has become the twelfth company to surpass a $1 trillion valuation, following a more than 180% increase in its stock this year. On the application front, Meta's Muse agent has surpassed ChatGPT to reach the top spot on the US App Store, coinciding with the group's best stock month in 13 years. However, limitations remain: Muse has a zero-day vulnerability, and Amazon has blocked the agent from making purchases on its platform. In consumer electronics, Dyson's $500 toothbrush is facing an issue related to an unspecified component.
Health: New Approaches from Lab to Human Trials
A battery designed to be ingested has been tested on pigs, combined with an RFID tracker and a gastric stimulator; in the future, it could power sensors, cameras, or drug release capsules. Epigenetic editing has begun human trials against hepatitis B, using a treatment based on chemical modifications to inactivate the virus's DNA. On the institutional front, the person in charge of the budget under Donald Trump may be granted veto power over NIH funding, the main global funder of biomedical research. In this economic context, Ben Casselman notes that developments in AI are concentrating contradictory concerns around work and growth.
Hype, Incentives, and Caution: The Perspective of Gebru and Bender
Timnit Gebru, who leads the Distributed AI Research Institute (DAIR), and Emily M. Bender, a linguistics researcher at the University of Washington, are sounding the alarm about the current overhype of AI. They point out that companies have recently ramped up announcements about breakthroughs, mathematical advancements, or the prospect of self-improving superintelligence, often accompanied by extensive media coverage. According to them, once the facts are examined by experts, different versions emerge. Gebru and Bender believe there is a commercial interest in exaggerating AI capabilities and that the urgency displayed by companies also aims to divert the attention of decision-makers and the public. They call for skepticism in the face of this hype.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.