AI
AI News

Latest AI News

Massive 31.4 Tbps DDoS attack breaks records: How the 'apex' of botnets could be weaponizing your home devices

Massive 31.4 Tbps DDoS attack breaks records: How the 'apex' of botnets could be weaponizing your home devices

Aisuru has set a new benchmark in cyber threats with a staggering 31.4 Tbps DDoS attack, surpassing previous records. This escalation emphasizes the increasing capabilities of cybercriminals and the urgent need for enhanced security measures among organizations. The implications for network providers and businesses are significant, highlighting vulnerabilities that must be addressed.

ZDNet
Coalition demands federal Grok ban over nonconsensual sexual content | TechCrunch

Coalition demands federal Grok ban over nonconsensual sexual content | TechCrunch

A coalition of nonprofits is calling for an immediate halt to the deployment of Grok, the AI chatbot created by Elon Musk’s xAI, within U.S. federal agencies, notably the Department of Defense. They express concerns over privacy, misinformation, and ethical implications of using AI in sensitive governmental operations.

TechCrunch
Docusign’s CEO on the dangers of trusting AI to read, and write, your contracts

Docusign’s CEO on the dangers of trusting AI to read, and write, your contracts

Allan Thygesen, CEO of DocuSign, emphasized that not offering AI services is no longer a viable option for the company. The statement reflects a growing trend among tech firms to integrate AI into their offerings, suggesting that DocuSign is likely to enhance its document management solutions with AI capabilities to stay competitive.

The Verge
AI layoffs or ‘AI-washing’? | TechCrunch

AI layoffs or ‘AI-washing’? | TechCrunch

Recent layoffs in tech firms have sparked debate about whether companies are genuinely restructuring for AI efficiencies or merely using AI as a justification for workforce reductions. Observers note that while some companies are implementing AI to enhance productivity and streamline operations, others appear to be making cuts without a clear strategy for integrating AI effectively. This distinction raises questions about the long-term sustainability of such layoffs and the true motivations behind them. As AI continues to evolve, companies must balance workforce adaptation with ethical considerations and employee retention strategies.

TechCrunch
India offers zero taxes through 2047 to lure global AI workloads | TechCrunch

India offers zero taxes through 2047 to lure global AI workloads | TechCrunch

India is incentivizing foreign cloud providers with a tax exemption on services sold outside the country until 2047, provided they host workloads domestically. This initiative aims to bolster India’s AI infrastructure and attract global tech investments, positioning the country as a competitive hub in the AI landscape.

TechCrunch
SpaceX seeks federal approval to launch 1 million solar-powered satellite data centers | TechCrunch

SpaceX seeks federal approval to launch 1 million solar-powered satellite data centers | TechCrunch

SpaceX has submitted a proposal to the FCC to deploy up to 1 million solar-powered satellites, which will function as data centers for AI applications. This initiative aims to enhance data processing capabilities globally. The project’s scale is unprecedented, potentially transforming how AI services are delivered and improving connectivity in underserved regions. Approval could significantly advance SpaceX's role in the satellite internet market, impacting competitors and the broader tech landscape.

TechCrunch
Physical Intelligence, Stripe veteran Lachy Groom's latest bet, is building Silicon Valley's buzziest robot brains | TechCrunch

Physical Intelligence, Stripe veteran Lachy Groom's latest bet, is building Silicon Valley's buzziest robot brains | TechCrunch

Physical Intelligence, located in San Francisco, operates discreetly, marked only by a subtle pi symbol on its entrance. The company focuses on developing advanced robotics and AI technologies aimed at enhancing human physical capabilities. Their innovations could significantly impact industries such as healthcare and manufacturing, by optimizing human-robot collaboration and improving efficiency in physically demanding tasks.

TechCrunch
Clawdbot, Moltbot, OpenClaw? The Wild Ride of This Viral AI Agent

Clawdbot, Moltbot, OpenClaw? The Wild Ride of This Viral AI Agent

Clawdbot, an open-source AI assistant designed to perform tasks on users' computers, went from viral sensation to controversy in just five days. The rapid rise included two rebrandings after backlash over its capabilities and perceived limitations. The situation highlights the challenges of managing user expectations for AI tools.

CNET
Bye Bye, ChatGPT-4o: OpenAI Will Soon Retire Several Models, Pushing New GPT-5 Ones

Bye Bye, ChatGPT-4o: OpenAI Will Soon Retire Several Models, Pushing New GPT-5 Ones

OpenAI announced it will retire several older models, including GPT-5, GPT-4o, GPT-4.1, GPT-4.1 mini, and o4-mini. The shutdown will take effect on a specified date, emphasizing a shift towards newer models. Users relying on these versions should transition to up-to-date alternatives to ensure continued performance and access.

CNET
Decoupled Diffusion Sampling for Inverse Problems on Function Spaces

Decoupled Diffusion Sampling for Inverse Problems on Function Spaces

The Decoupled Diffusion Inverse Solver (DDIS) introduces a novel framework for solving inverse PDE problems, emphasizing data efficiency and physics-awareness. Unlike traditional models that require extensive paired data, DDIS separates coefficient and solution learning, leading to an 11% improvement in $l_2$ error and a 54% reduction in spectral error under sparse data conditions. Notably, when trained with just 1% of available data, DDIS outperforms joint models by 40% in $l_2$ error, showcasing its effectiveness in low-data scenarios.

arXiv
IRL-DAL: Safe and Adaptive Trajectory Planning for Autonomous Driving via Energy-Guided Diffusion Models

IRL-DAL: Safe and Adaptive Trajectory Planning for Autonomous Driving via Energy-Guided Diffusion Models

A new framework, IRL-DAL, enhances autonomous vehicle navigation through a diffusion-based adaptive lookahead planner. It begins with imitation learning from an expert FSM controller, integrating environmental feedback and IRL rewards. The approach achieves a 96% success rate and reduces collisions to 0.05 per 1,000 steps, significantly improving safety. The code is publicly available for further research.

arXiv