AI news and funding updates from the last 24 hours(2nd April 2025):

✅ Google DeepMind

- AGI Safety Research:

Google DeepMind has released a paper that outlines its safety approach to Artificial General Intelligence (AGI). The research discusses potential risks associated with recursive AI improvement and proposes mitigation techniques, sparking debate about the realistic timeline and challenges of achieving AGI.

✅ Cognizant

- Leadership Appointment & Strategy:

Sailaja Josyula has been named Global Head of Cognizant's Global Capability Center Service Line in Hyderabad. She will lead the worldwide GCC strategy, leveraging her extensive leadership experience—including her prior role heading Cognizant’s Hyderabad operations.

- Training Initiative:

Cognizant is set to build a Chennai-based learning center that will annually train 100,000 individuals in AI.

✅ Microsoft

- Cybersecurity Enhancement:

Microsoft Security Copilot, an AI-powered tool, has identified vulnerabilities in open-source bootloaders such as GRUB2, U-Boot, and Barebox. The tool expedited vulnerability detection, prompting the release of patches to address potential security risks.

✅ Meta

- Executive Departure:

Joelle Pineau, Meta's VP of AI Research, has announced her resignation. Having led Meta’s FAIR research lab and overseen major open-source initiatives like PyTorch and Llama, Pineau’s departure marks a significant transition as Meta ramps up its AI investments and restructures its AI research efforts.

✅ RRP Electronics

- Semiconductor Exports:

Mumbai-based RRP Electronics became the first Indian company to export packaged semiconductors under the government's Semiconductor Mission. Their initial export focused on ASICs, supporting the “Make in India” initiative, while the company continues to expand through strategic partnerships and new ventures.

✅ Emergence AI

- Platform Upgrade:

Emergence AI has enhanced its platform to enable the autonomous creation and assembly of AI agents. The upgraded system features an “orchestrator” that automates coding, planning, and management of AI agents, reducing human intervention while allowing oversight.

✅ OpenAI

- Educational Expansion:

OpenAI has broadened its Academy by launching a free online resource hub. This initiative provides tutorials, courses, and events designed to boost AI literacy for educators, students, and business owners globally.

- Copyright Allegations:

A new paper from the AI Disclosures Project alleges that OpenAI’s GPT-4o model was trained using unlicensed, copyrighted books, raising questions about its data sourcing practices.

✅ Wikimedia Foundation

- Infrastructure Strain:

Wikimedia Foundation reports a 50% increase in multimedia bandwidth consumption since January. The surge is attributed to AI scrapers disproportionately accessing content, prompting efforts to block these automated crawlers to preserve resources for human users.

✅ Parasail

- GPU Access Platform:

Parasail, an AI startup, is launching a platform that provides on-demand GPU access. Aimed at challenging cloud giants like AWS and Azure, the service will offer high-performance GPUs (including Nvidia’s H100) at competitive prices to support the growing demand for AI infrastructure.

✅ Anthropic

- Education-Focused Chatbot:

Anthropic has introduced Claude for Education, a new tier of its AI chatbot tailored for higher education. This version includes features such as a “Learning Mode” and is designed to enhance critical thinking and streamline administrative processes, in direct competition with similar offerings from OpenAI.

✅ Actively AI

- Sales Intelligence Innovation:

Actively AI, a startup in the AI sales rep space, leverages advanced “reasoning” models to identify high-value sales prospects. Their custom solution, which integrates models from OpenAI and Anthropic, has already generated significant additional revenue for clients like Ramp, attracting strong investor support.