AI News

Follow the latest AI and automation news

4 Jul 202619 Muharram 1448 AH

Midjourney wants Hollywood studios to reveal the details of their AI usage

Midjourney is currently engaged in a legal dispute with three Hollywood studios, seeking to compel them to disclose their AI usage. This case raises important questions about intellectual property rights and competitive practices in the industry, particularly regarding how these studios leverage AI in content production. If Midjourney succeeds in this effort, it could significantly reshape industry standards and practices around AI technology usage.

Governance

4 Jul 202619 Muharram 1448 AH

Anthropic Implements Strategy to Counter Use of Its Tools in China

Anthropic has launched an aggressive operational plan to close loopholes that allowed Chinese companies to access its tools. Entities like Ant Financial have utilized indirect channels, including international cloud services, to reach Anthropic's offerings. As one of the strictest AI companies in the Chinese market, Anthropic enforces rigorous user verification and prohibits payments from Chinese banks.

Models

4 Jul 202619 Muharram 1448 AH

What is Mistral AI? Everything to know about the OpenAI competitor

Mistral AI has raised significant funding since its establishment in 2023, aiming to make advanced AI accessible to everyone. This move reflects a growing trend towards open-source AI models, which are increasingly sought after for their collaborative potential. The funding will enable Mistral AI to enhance its models and improve performance, positioning the company competitively in the market. This aligns with the broader shift towards democratizing AI technology, allowing more developers and users to engage with these tools.

💡

Global Energy Challenges from Increased AI Data Center Consumption

What strategies can balance energy consumption with technological growth?

Models

Nobel-winning materials scientist Omar Yaghi joins China’s Tsinghua University from the US

4 Jul 202619 Muharram 1448 AH

Models

4 Jul 202619 Muharram 1448 AH

Nobel-winning materials scientist Omar Yaghi joins China’s Tsinghua University from the US

Omar Yaghi, last year's Nobel Prize winner in chemistry, has left the U.S. to lead a new AI-driven research center at Tsinghua University. At 61, he will oversee a team focused on leveraging AI to transform the design and synthesis of new materials, aiming to significantly shorten their development cycle. Tsinghua University aims to harness AI technologies to enhance research and development efficiency in materials science. The team seeks to achieve notable advancements in material design, which could lead to innovations in energy and environmental sectors.

Models

Fu Cong Team and Xiamen University Propose ManCAR: Manifold-Constrained Adaptive Reasoning Boosts Recommendation by 46%

4 Jul 202619 Muharram 1448 AH

Models

4 Jul 202619 Muharram 1448 AH

Fu Cong Team and Xiamen University Propose ManCAR: Manifold-Constrained Adaptive Reasoning Boosts Recommendation by 46%

AI-driven recommendation systems are evolving from traditional scoring models to multi-step reasoning approaches. Fu Cong's team, in collaboration with Xiamen University, introduced the ManCAR framework, accepted at KDD 2026, which enhances generative recommendations by achieving a 46.88% improvement in NDCG@10, as reported in their paper on InfoQ.

Infrastructure

From Gigabit to 10-Gigabit: Huawei AI-FAN Architecture Drives Broadband Home Upgrade

4 Jul 202619 Muharram 1448 AH

Infrastructure

4 Jul 202619 Muharram 1448 AH

From Gigabit to 10-Gigabit: Huawei AI-FAN Architecture Drives Broadband Home Upgrade

Huawei has unveiled its AI-FAN architecture, a new generation of optical access networks, at MWC Shanghai 2026. This framework aims to facilitate the shift from gigabit to 10-gigabit home broadband, with 32 million 10G PON ports deployed in China by February 2026. The AI-FAN architecture embodies Huawei's vision for AI-driven home network transformation, focusing on user experience through Wi-Fi optimization and network slicing. Over 90% of provincial operators have launched gigabit-plus packages, with more than 500 million subscribers benefiting from these services.

Infrastructure

Semiconductor Prices Skyrocket as AI Demand Fuels Industry-Wide Chip Shortage

4 Jul 202619 Muharram 1448 AH

Infrastructure

4 Jul 202619 Muharram 1448 AH

Semiconductor Prices Skyrocket as AI Demand Fuels Industry-Wide Chip Shortage

On July 1, 2026, a significant surge in semiconductor prices began across the global chip industry, affecting over 20 major manufacturers. AI-specific multilayer ceramic capacitors saw price increases of 3 to 10 times, disrupting the pricing system. The unprecedented supply crunch is driven by rising raw material costs and capacity constraints, with demand for AI components skyrocketing.

Agents

OpenAI cofounder envisions "almost no interface" future where nobody learns software anymore

4 Jul 202619 Muharram 1448 AH

Agents

4 Jul 202619 Muharram 1448 AH

OpenAI cofounder envisions "almost no interface" future where nobody learns software anymore

Greg Brockman, co-founder of OpenAI, acknowledged that the ChatGPT plugins heavily marketed in 2023 failed due to the models' lack of readiness. Instead of relying on plugin applications, Brockman envisions a future with an invisible, intelligent agent that interacts contextually. However, OpenAI's Codex model is still far from realizing this future vision. Brockman's idea of a smart agent without a visible interface represents a bold step towards the future of software, where people are expected to learn less about using software. This shift could transform our interaction with technology, making it smoother and more efficient.

Models

Mistral's open-source Leanstral 1.5 aces formal math benchmarks and catches real bugs in code

4 Jul 202619 Muharram 1448 AH

Models

4 Jul 202619 Muharram 1448 AH

Mistral's open-source Leanstral 1.5 aces formal math benchmarks and catches real bugs in code

Mistral AI has launched Leanstral 1.5, an open-source model for formal verification using Lean 4. During an examination of 57 open-source repositories, the model identified five previously unknown bugs, demonstrating its high efficiency. Leanstral 1.5 represents a significant advancement in formal verification, offering sophisticated mechanisms for analyzing mathematical and programming code.