AI News

DigitalOcean is enhancing its cloud services for small and mid-sized businesses (SMBs) by offering support for Nvidia H100 GPUs via its Paperspace platform, allowing scalable access to advanced AI capabilities. This aims to lower barriers for SMBs, enabling them to develop various AI applications efficiently.

Databricks launched its Data Intelligence Platform for Communications, tailored for telecom and network service providers. Leveraging advanced AI and data lakehouse architecture, the platform streamlines workflows, enhances customer insights, and drives growth, already benefiting companies like AT&T and T-Mobile.

Generative AI models from companies like OpenAI and Meta are facing backlash for using scraped data without creator consent, leading to lawsuits from authors and artists. A new nonprofit, "Fairly Trained," advocates for obtaining explicit consent and has introduced a certification for companies that respect creators' rights.

DataStax has launched a new data API to simplify the development of generative AI RAG applications, enabling developers to use Python and JavaScript with AstraDB. This API streamlines interactions, enhances vector data performance, and supports JVector for improved retrieval, positioning DataStax competitively in the market.

Google DeepMind's AlphaGeometry, an AI system, can solve complex geometry problems at a level on par with human IMO gold medalists. Combining a neural language model with a symbolic deduction engine, it solved 25 of 30 challenging problems, showcasing AI's potential for mathematical reasoning and discovery.

Samsung and Google Cloud have partnered to integrate generative AI into the Galaxy S24 series, enhancing communication and creativity with tools like Gemini Pro and Imagen 2. The series features advanced AI-driven functions, such as real-time translation and image editing. Pre-orders start January 17, 2024.

At the World Economic Forum, CEO Sam Altman expressed a tempered view on artificial general intelligence (AGI), stating its meaning is unclear and its impacts may be less dramatic than anticipated. He emphasized ongoing improvements rather than a sudden breakthrough, reflecting a shift from his previous alarmist stance. The OpenAI board's restructured dynamics also influence these discussions.

Udacity has launched a Generative AI Nanodegree Program to equip professionals with essential skills for utilizing generative AI across industries. The program features practical projects, mentorship, and is accessible to All Access subscribers. Prerequisites include intermediate Python and SQL skills.

Pecan AI has launched Predictive GenAI, combining generative AI and predictive machine learning to make predictive analytics accessible to businesses without deep technical expertise. With features like Predictive Chat and Predictive Notebook, it simplifies model creation and data preparation while addressing challenges like data leakage.

New York's Vicarius has raised $30 million in Series B funding to enhance its AI-powered vulnerability management platform, vRx. With over 400 clients, including major corporations, Vicarius aims to automate vulnerability remediation. Future plans include the x_comply compliance tool and improvements to vuln_GPT, which generates remediation scripts using LLM technology.

A 2023 survey of 2,778 AI experts reveals rapid advancements in AI capabilities, with a 50% chance machines could outperform humans in all tasks by 2047. While concerns over risks like job displacement and potential human extinction abound, 68% believe AI's benefits will outweigh negatives.

Toyota has partnered with Ready Robotics to enhance industrial robotics using Nvidia's Omniverse platform, integrating simulation and programming for its hot forging production lines. This collaboration enables a "sim-to-real" workflow, improving efficiency, safety, and automation through digital twins and real-time data feedback.

Finally Secures $50 Million in Series B Funding to Enhance SMB Financial Services

During a significant road trip across the U.S. years ago, Sanish Mondkar observed a troubling disconnect between employers and their employees. To some critics of late-stage capitalism, this may seem like a familiar critique. However, for Mondkar, who holds a master’s degree in computer science from Cornell, witnessing these challenges firsthand was a pivotal moment.

Facebook has advanced speech recognition by developing Wav2vec-U, an "unsupervised" model that learns from audio and unpaired text, eliminating the need for transcribed data. This innovation improves accuracy across diverse languages and could enhance global access to technology, supporting Facebook's mission to connect billions.