SCRPika Labs has unveiled a new version of its AI video generation model. Pika 1.5 offers spectacular special effects that defy the laws of physics, reports Carl Franzen. The software allows users in its videos to melt, crush, or transform objects into cakes. Realistic movements and cinematic camera shots are now also possible. The basic …
DeepMind has developed a new technique called SCoRe that significantly improves the self-correction abilities of large language models (LLMs). Ben Dickson reports this in an article for VentureBeat. SCoRe uses self-generated data and enables LLMs to use their internal knowledge to identify and correct errors. In tests, SCoRe significantly outperformed other self-correction methods. The technique …
Nvidia has released a powerful open-source AI model that rivals proprietary systems from industry leaders like OpenAI and Google. The model, called NVLM 1.0, demonstrates exceptional performance in vision and language tasks while also enhancing text-only capabilities. Michael Nuñez reports on this development for VentureBeat. The main model, NVLM-D-72B, with 72 billion parameters, can process …
Mostly AI has introduced a new feature for generating synthetic texts. The tool allows companies to use confidential data such as emails and conversations for AI training without privacy concerns. As reported by Shubham Sharma, the platform generates a version of proprietary information free from personally identifiable data. This enables businesses to train and optimize …
A new AI service enables real-time conversations with digital twins at a latency of less than one second. Startup Tavus has developed this technology, as founders Hassaan and Quinn report on HackerNews. Their goal was to enable the most natural communication possible between humans and computers. Numerous technical challenges had to be overcome to achieve …
A data scientist from Singapore has developed an open-source alternative to Google’s NotebookLM. Gabriel Chua of the GovTech agency created the tool called “Open NotebookLM” in just one afternoon. It converts PDF documents into personalized podcasts using publicly available AI models. The project demonstrates how quickly complex AI applications can be replicated today. However, the …
Series Entertainment has closed a $28 million funding round, according to an SEC document and confirmation from the company, as reported by Julie Bort for TechCrunch. Investors include Netflix, Dell Technologies Capital, and existing backers Andreessen Horowitz, BITKRAFT, and F4 Fund. Series is developing an AI-powered platform for game developers called the Rho Engine. The …
Startup Artisan has raised $11.5 million in seed funding to develop AI-powered “employees” for sales teams. The company offers an AI assistant named Ava that automates tasks such as lead research and personalized email creation. CEO Jaspar Carmichael-Jack told VentureBeat that Artisan has already secured over 120 customers and reached $1 million in annual revenue. …
Character.ai is changing its strategy following a $2.7 billion deal with Google. As new interim CEO Dominic Perella told the Financial Times, the company is abandoning the costly development of its own AI models. Instead, Character.ai will focus on improving its popular chatbot products that simulate conversations with various characters and celebrities. After the Google …
AI chipmaker Cerebras Systems has filed for an initial public offering on the Nasdaq. The company plans to trade under the ticker symbol “CBRS,” according to the prospectus. Cerebras competes with Nvidia in the AI chip market, claiming its WSE-3 chip has more cores and memory than Nvidia’s popular H100. In the first six months …