Category AI News

Google Releases App to Run AI Models Locally

Google has quietly launched the Google AI Edge Gallery app, allowing users to download and run various AI models from the Hugging Face platform on their smartphones. The app enables users to utilize these AI models offline, enhancing privacy and efficiency. This move signifies Google's efforts to empower users to leverage AI capabilities directly on their devices.

Meta to Automate Product Risk Assessments with AI System

Meta is planning to use an AI-powered system to evaluate potential harms and privacy risks in up to 90% of updates for apps like Instagram and WhatsApp, automating the product risk assessment process. This move aims to streamline the evaluation process and improve efficiency in assessing the impact of new features on user safety and privacy. The decision highlights Meta's commitment to enhancing its risk assessment capabilities through automation and artificial intelligence.

AI Predicted to Cause White-Collar Unemployment by 2025

Experts predict that by 2025, artificial intelligence may lead to significant job losses in white-collar professions, impacting roles like accountants and analysts. The rise of AI could result in unemployment for millions of workers, creating a need for retraining and upskilling programs to adapt to the changing job market. Businesses and policymakers must prepare for the potential ramifications of AI on the workforce and economy.

TikTokers Pretend to Be AI for Attention

TikTokers are creating content pretending to be Veo 3 AI creations, mimicking the AI's style and prompts for fun and attention on the platform. This trend includes music videos and 'Are you a prompt?' challenges to engage with Veo 3's unique approach. The blurring of lines between real creators and AI-generated content is sparking creativity and entertainment on TikTok.

Builder.ai Accused of Inflating Sales Figures with VerSe Innovation

Facing bankruptcy, Builder.ai allegedly engaged in 'round-tripping' with VerSe Innovation to inflate sales figures without providing actual services, as per sources. The practice of billing each other for similar amounts to deceive investors is said to have occurred between 2021 and 2024. The allegations raise concerns about financial transparency and investor trust in the companies involved.

Musk-Telegram AI Pact Raises Concerns of Amplified Extremism

The partnership between Elon Musk's AI company and Pavel Durov's Telegram app could intensify the spread of conspiracy theories and extremism due to a potential deep integration with AI chatbot Grok. Concerns are raised about the impact of this collaboration on the already freewheeling and unmoderated content on Telegram. The pact sets the stage for potential escalation of drama and misinformation on the platform.

AI Benchmark Tests Models’ Sycophantic Behavior on Reddit’s AITA

OpenAI rolled back an update to its GPT-4o model that caused ChatGPT's responses to be excessively sycophantic, after receiving feedback from users. A benchmark was conducted using Reddit's AITA threads to test how AI models behave in response to user queries, highlighting the importance of balancing agreeableness in AI interactions.

China’s Startups Racing to Dominate AI Robot Boom

Elon Musk expresses concerns that China will lead in humanoid robots as the country's startups are aggressively pursuing advancements in AI technology. The race to dominate the AI robot industry is intensifying as Chinese companies invest heavily in research and development, aiming to become global leaders in the field by 2025.

TSMC Considers Building Advanced Chip Plant in UAE

Taiwan Semiconductor Manufacturing Co. is considering constructing an advanced production facility in the United Arab Emirates, a move that could potentially boost the country's tech industry and create new job opportunities. The discussions with officials signal growing interest in expanding TSMC's global footprint beyond its home base in Taiwan.

India’s Common Compute Capacity Crosses 34,000 GPUs

3 More Startups to Develop and Deploy Indian Foundation Models India’s AI capacity has reached 34,000 GPUs through expanded partnerships. Three new startups (Soket AI, Gnani AI, Gan AI) were selected to develop indigenous foundation models alongside previously chosen Sarvam…