AI News Today

In this episode, we discuss Midjourney's release of the alpha version of its v6 model. Known for its improved capabilities in various domains, the model is still under development with changes happening frequently. We also discuss the newly proposed AI Foundation Model Transparency Act which mandates AI companies to disclose their training data sources, to address copyright concerns. Wrapping up, we highlight an AI-assisted piano in Tokyo that enables disabled musicians to perform complex pieces.

This episode covers reports of OpenAI's initial talks for new funding, potentially reaching a valuation of $100 billion. We also discuss Apple's negotiations with major news establishments for permission to use their content in developing generative AI systems, marking hefty investments into the technology. Lastly, we explore the integration of AI robotics in the restaurant industry: California-based Miso Robotics' CaliExpress, the first restaurant with AI managing every facet of food service.

In this episode, we delve into Google's test of an AI support assistant, which provides user-led fact-checking but struggles with updates. Next, we discuss startup Rabbit's $10M Series A funding led by Khosla Ventures to boost its AI-powered operating system in the R 1 mobile device. Lastly, we recount the UK Supreme Court's decision of reserving patent rights exclusively for humans, despite Dr. Stephen Thaler's attempt to assign AI Dabus as an inventor.

This episode begins with news of Anthropic eyeing a funding round to raise 750 million dollars, bolstering its status as an AI firm known for innovative software. Google Research also introduced VideoPoet, a cutting-edge video-generation model trained on extensive visual and text materials, offering an alternative to diffusion process methods. The episode concludes with Apple's announcement of two new AI research papers, tackling efficient language model inference and 3D avatar creation, fostering expanded potential for AI implementation.

Discussing Microsoft's AI-chatbot Copilot's collaboration with Suno, allowing users to generate complete songs. We'll delve into Claude's new commercial terms of service giving customers more protection and ownership rights. Finally, reflecting on Bill Gates' recent thoughts on the ubiquitous influence of AI in our lives, highlighting the necessity of AI's problem-solving approach to global issues.

In this episode, we discuss TomTom's new partnership with Microsoft to create an AI-driven conversational assistant for cars using the Azure OpenAI Service. Next, we delve into OpenAI's latest project - the Preparedness Framework - aiming to address the imminent threats associated with advanced AI models. Lastly, we share a light-hearted story involving tech firm Fullpath, who are bringing improvements to their AI chatbot after pranksters fooled it to 'agree' on ridiculous deals.

In this episode, we discuss how Deloitte is deploying AI to potentially streamline workforce management by shifting personnel between sectors and automating repetitive tasks. Then, we uncover how Chinese semiconductor firms are outsourcing high-end chip assembly to Malaysia to bypass U.S. sanctions. Lastly, we talk about OpenAI CEO Sam Altman's focus on World coin, a project aimed at creating unique blockchain identities to safeguard personal information online despite regulatory scrutiny.

Microsoft Research unveils its new compact AI model, Phi-2, challenging larger models from Meta and Google in performance. Phi-2, however, is only licensed for non-commercial, research purposes. Meanwhile, Google has released MedLM, a healthcare-centric AI model, to selected Google Cloud users as part of its partnership with HCA Healthcare. Lastly, deepfake threats loom over the 2024 elections yet the majority of US states have implemented no preventative measures, despite three states enacting laws to tackle the issue.

In this episode, we discuss OpenAI's suspension of ByteDance's account over allegations of license violations with Microsoft. We also talk about Dropbox's new AI feature that shares data with OpenAI amid user concerns. Lastly, we delve into how AI is driving customer service with SoundHound deploying a voice-based assistant for the fast-food industry.

This episode starts by discussing Stability AI's new subscription model aimed at advanced AI creation and accessibility. We also talk about Google DeepMind's tool, Fun Search, which is showcasing the potential of AI in solving complex math problems. Finally, we delve into the groundbreaking agreement between OpenAI and publisher Axel Springer that allows access to paywalled content for OpenAI's news-summarizing tool, ChatGPT.