Nvidia's New LLM; AI Integrity Suite; OpenAI's Massive Funding & Canvas; Funds for MLOps, Customer Support, AI Assistant, Autonomous Engineer ...

Zhitao's Top Picks of the Week - News (03/10/2024 - 09/10/2024)

1. Nvidia's New Open-Source LLM: NVIDIA has released a 72 billion-parameter open-source large language model (LLM) named NVLM-D-72B. This model is part of the NVLM 1.0 family and is designed to perform on par with leading proprietary models from companies like OpenAI and Google. Nvidia's approach includes making the model's training weights available for public review, positioning it as a foundation for third-party developers to build their own AI applications. READ
2. AI Integrity Suite for Transparent AI Models: Abu Dhabi wealth fund ADQ has partnered with EQTY Lab to integrate an AI Integrity suite across its portfolio, ensuring reliable and transparent AI models. This collaboration previously launched ClimateGPT, an AI model focused on climate change, hosted at the Al Dhafra Solar PV plant. ADQ's portfolio companies like ADNEC Group and Masdar (Abu Dhabi Future Energy Company) will adopt ClimateGPT. AI is projected to significantly boost the Middle East’s GDP, especially in construction and energy sectors. READ (Zhitao: this can be the first of its kind in production)

Investments

1. MLOps Platform: VESSL AI , a South Korean MLOps platform, has raised $12 million to enhance its infrastructure aimed at reducing GPU costs by up to 80%. The platform uses a hybrid infrastructure model to optimise GPU expenses and streamline AI model development. READ
2. Customer Service Workflows: AI startup Lorikeet has emerged from stealth, raising $5M from Square Peg and angel investors. Founded by ex-Stripe's Steve Hind and ex-Google's Jamie Hall, Lorikeet aims to automate complex customer service workflows. Competing with giants like OpenAI and Sierra, Lorikeet's first client is telehealth company Eucalyptus. READ
3. AI Assistant: kapa.ai, a startup transforming technical knowledge bases into AI assistants, raised $3.2M in seed funding led by Initialized Capital and backed by Y Combinator and notable angels like Douwe Kiela, Amjad Masad and Solomon Hykes. Co-founded by Emil Sorensen and Finn Bauer, Kapa.ai supports over 100 companies, including Docker, OpenAI, and Monday.com, offering explainable, secure, and up-to-date AI-driven answers with extensive integrations. READ
4. Enterprise Data Analytics: e6data, an enterprise data analytics firm, secured $10M in Series A funding led by Accel with participation from BEENEXT. Co-founded by Vishnu Vasanth, e6data's platform offers high-efficiency, format-neutral compute engines to enhance ROI, eliminate data movement, and reduce downtime. The company works with Fortune 500 customers and partners like Chargebee. The data and AI market is projected to reach $230B by 2025. READ
5. Autonomous Software Engineers: Lovable, a Stockholm-based startup developing autonomous software engineers, raised $7.5M in pre-seed funding led by Hummingbird and byFounders. The round included notable angels like Mattias Miksche and Siavash Ghorbani. Co-founded by Anton Osika and Fabian Hedin, Lovable's platform uses AI to generate and maintain production-ready code based on user input. With over 27,000 on its waitlist and 52,000 GitHub stars, the company aims to scale its team and enhance its technology. READ

News

1. OpenAI's Massive Funding and Valuation: OpenAI has closed a $6.6 billion funding round, raising its valuation to a staggering $157 billion. This round included investments from major players like Microsoft and NVIDIA . The company is also considering an IPO to further capitalise on its growth and secure additional capital for its ambitious goals. READ
2. OpenAI's New Canvas Tool: OpenAI has introduced a new feature called Canvas, designed to enhance collaboration in writing and coding projects using ChatGPT. This tool allows users to view and edit projects more efficiently, providing shortcuts for writing and coding tasks. READ
3. AI in Healthcare: AI's role in healthcare continues to expand, with companies like Google 's DeepMind and BioNTech SE developing AI-powered lab assistants. However, there are concerns about biases in AI models, which could have negative implications for patient care. READ

Stay tuned for more updates in next week’s digest!

Want to learn more about how we curate the latest AI digest or get your hands dirty on the analysis? Reach out to discover how Tangible AI Digest is created and stay connected with exclusive content, discussions, and updates. Subscribe now to stay ahead in the AI revolution!