AI News

Follow the latest AI and automation news

12 Sept 20261 Rabiʻ II 1448 AH

Growing Calls for AI Regulation Amid Rising Concerns

Concerns are rising regarding the capabilities of AI models, prompting calls for government intervention to regulate the sector. Writer Robert Sirin suggests that the pharmaceutical industry could provide a model for managing these risks, as regulations evolved after past incidents. These calls come as Sam Altman, CEO of OpenAI, discusses the possibility of slowing down the development of advanced models. Jacob Coxon, a former researcher at Anthropic, resigned due to fears about the risks associated with developing self-improving models.

Governance

12 Sept 20261 Rabiʻ II 1448 AH

Anthropic CEO outlines plan to slow AI development

Dario Amodei from Anthropic and Sam Altman from OpenAI discussed the necessity of regulating AI developments. They emphasized the importance of setting clear boundaries to guide innovations responsibly, reflecting their awareness of ethical and technical challenges. This approach marks a significant step in the AI landscape, as leaders in the field seek to balance innovation with safety. The emphasis on collaboration among companies to develop common standards was highlighted.

Governance

12 Sept 20261 Rabiʻ II 1448 AH

Anthropic CEO says it’s time to pump the brakes on AI

Dario Amodei, CEO of Anthropic, has announced the need to slow down AI development. He proposed a three-step plan aimed at pacing training and development, allowing companies time to implement safety measures. The first step involves granting external evaluators like METR access to Anthropic's models to ensure adherence to safety practices. Amodei views this as a unilateral initial action that sets the stage for broader industry collaboration.

💡

Call to Slow AI Model Development for Safety Assurance

What measures can ensure safety without stifling AI innovation?

Business

Concerns Over AI Boom Sustainability Amid Funding Gaps

12 Sept 20261 Rabiʻ II 1448 AH

Business

12 Sept 20261 Rabiʻ II 1448 AH

Concerns Over AI Boom Sustainability Amid Funding Gaps

Concerns are rising about the sustainability of the AI boom as tech companies invest heavily in infrastructure, currently spending over a trillion dollars annually while generating only around 200 billion dollars in revenue. Analyses indicate that the gap between spending and revenue means AI companies require substantial funding sources to continue building data centers and expanding capabilities. As capital needs grow, the sector increasingly relies on stock and bond markets.

Governance

New Legislation in US Congress to Regulate Artificial Intelligence

12 Sept 20261 Rabiʻ II 1448 AH

Governance

12 Sept 20261 Rabiʻ II 1448 AH

New Legislation in US Congress to Regulate Artificial Intelligence

Negotiations in the US Senate are considering new legislation that would hold tech companies accountable for designing safer AI systems amid rising concerns over misuse. Lawmakers are discussing a bill that would impose a 'duty of care' on developers to mitigate risks, including the potential use of AI systems in developing nuclear weapons.

Governance

Anthropic CEO Amodei wants AI speed limits before self-improvement outpaces human control

12 Sept 20261 Rabiʻ II 1448 AH

Governance

12 Sept 20261 Rabiʻ II 1448 AH

Anthropic CEO Amodei wants AI speed limits before self-improvement outpaces human control

Dario Amodei, CEO of Anthropic, is advocating for a controlled slowdown in AI development. He warns that recursive self-improvement could threaten the entire internet within six to twelve months. Amodei proposes implementing embedded auditors at AI companies, establishing shared safety standards, and creating global agreements modeled after the SALT disarmament treaties. His warning comes just before what could be the largest initial public offering in history.

Agents

GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks

12 Sept 20261 Rabiʻ II 1448 AH

Agents

12 Sept 20261 Rabiʻ II 1448 AH

GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks

GPT-6 Astra has achieved significant advancements in spatial understanding according to new robotics benchmarks. On the StationeryBench platform, the model completed 7 out of 100 tasks using dual-arm robots, while competitor MolmoAct2 failed to finish any tasks. This achievement is considered a 'step change' in spatial reasoning. The results indicate that GPT-6 Astra significantly outperforms its competitors in spatial understanding tasks, reflecting substantial improvements in the model's capabilities. Researchers suggest that this progress could enhance practical applications across various fields, including intelligent robotics.

Business

Nvidia wants to pour up to $10 billion into Anthropic's record-breaking IPO

12 Sept 20261 Rabiʻ II 1448 AH

Business

12 Sept 20261 Rabiʻ II 1448 AH

Nvidia wants to pour up to $10 billion into Anthropic's record-breaking IPO

Nvidia is in talks to invest up to $10 billion in Anthropic's planned IPO, as reported by Reuters. If successful, Anthropic's target valuation would be $2 trillion, making it the largest IPO in history. Most of this investment is likely to return to Nvidia through chip orders. This potential investment represents a strategic move for Nvidia, enhancing its position in the AI market and reflecting the increasing demand for advanced AI technologies.

Models

AI models' written reasoning steps correspond to distinct internal patterns, a new study finds

12 Sept 20261 Rabiʻ II 1448 AH

Models

12 Sept 20261 Rabiʻ II 1448 AH

AI models' written reasoning steps correspond to distinct internal patterns, a new study finds

A new study reveals that reasoning steps in AI models, such as calculations and formula retrieval, can be distinctly separated in the model's internal states. The findings indicate that models process more information than their visible chain of thought suggests, highlighting the importance of understanding these patterns in the context of AI safety. This discovery could influence how AI models are developed, as developers need to consider these internal patterns to ensure the safety and reliability of systems.