AI Advancements and Industry Updates on OWITH.ai: Navigating the Latest in Tech News

04/18/25 • 3 min

OWITH.ai - Only What's Important to Hear around AI and Tech

Good morning from OWITH.ai: the podcast that gives you only what's important to hear in AI and tech world. In a recent critique, Tyler Cowen's extreme praise of OpenAI's O3 model as AGI is called into question by author Gary Marcus. Marcus points out flaws in the model highlighted by experts and urges for a more honest assessment of AI's strengths and weaknesses for improvement, emphasizing the need for a broader perspective on AI's capabilities.Moving on to other news, OpenAI has established a new nonprofit commission with notable advisors to enhance its credibility during its transition into a for-profit entity. This strategic move aims to address challenges from various organizations and individuals accusing the company of diverting charitable assets for private gain. The commission serves as a means to maintain support and credibility amidst this transition, highlighting the importance of transparency in AI development and governance.Former SpaceX engineer Scott Morton has recently launched Revel, a startup focused on building software to control critical hardware and prevent potential disasters. With $30 million in funding from investors like Thrive Capital and Felicis, Revel aims to revolutionize how industrial companies manage their technological infrastructure, starting with a focus on the space industry before expanding into other sectors like manufacturing and automotive.OpenAI has also introduced new "reasoning" models and a coding agent called Codex CLI, offering users access to advanced tools for reasoning about visual information. While Meta CEO Mark Zuckerberg warns of competitive threats from TikTok, Spotify faces widespread outages, highlighting the challenges in the tech industry. As AI-generated music rises in popularity, companies like NVIDIA and AMD anticipate higher costs amid trade war tensions, underscoring the competitive landscape and challenges faced by tech companies.Fed Chair Jerome Powell's warning about potential economic impacts from Trump's tariffs has global markets on edge, with U.S. stocks sliding amidst uncertainty. As trade tensions escalate, U.S. CEOs face challenges in navigating partnerships in Asia, with countries weighing their options between the U.S. and China in key sectors like EVs and AI. Amidst discussions about the dollar's reserve currency status, global markets brace for potential economic shifts driven by geopolitical uncertainties.The importance of building automated evaluations for AI applications is emphasized, with Google's Gemini 2.5 model and OpenAI's Model Context Protocol facilitating developers' access to tools and data. The development of Byte Latent Transformer (BLT) offers a robust solution for processing noisy inputs and unfamiliar languages, showcasing advancements in AI technology. As job openings in AI companies continue to grow, it is essential to stay updated on the latest developments to provide valuable feedback and insights into the evolving landscape of AI technology.In conclusion, OpenAI's recent releases of O3 and O4-mini models signify a shift towards integrated AI systems that combine perception, action, and reasoning. While these advancements are groundbreaking, further testing is necessary to fully grasp the capabilities of these new AI systems, marking a significant evolution in how we approach AI technology moving forward.

Thanks for listening! Follow us on Twitter, Instagram and Linkedin

Thanks for listening! Follow us on Twitter, Instagram and Linkedin

Previous Episode

AI Agents, Writing Struggles, Healthcare Startups, and More: The Latest in AI and Tech News from OWITH.ai

Good morning from OWITH.ai: the podcast that gives you only what's important to hear in AI and tech world.AI agents are not likely to replace most programmers anytime soon, despite claims to the contrary. While AI agents have the potential to change the world, they currently struggle with debugging and producing code that is free of bugs and security vulnerabilities. Most researchers agree that the best outcome is for AI agents to save human developers time rather than completely replacing them. A recent Microsoft study highlighted the challenges of getting AI to debug reliably, confirming that debugging is a crucial aspect of coding that AI struggles with. In order to improve AI agents, a paradigm shift is necessary. Gary Marcus emphasizes the importance of acknowledging the limitations of current AI technology and adjusting expectations accordingly.Moving on to the next news, the text discusses why AI is not improving in writing despite advancements in other areas. The author questions why writing, a craft considered inferior to STEM disciplines, remains difficult for AI. The trendline for AI in writing is described as a plateau, in contrast to the steep trendlines seen in other fields like chess. The author explores the reasons behind AI's struggle with writing and invites readers to subscribe to access more content on the topic.In other news, Chemistry's Kristina Shen, a former A16Z investor, has made her first investment in Assort Health, which raised $22 million in a Series A round led by First Round and Chemistry. Assort Health was founded by Jon Wang and Jeff Liu, who met through First Round's "cofounder dating" questionnaire and decided to combine their expertise in AI and medicine to create voice AI agents for healthcare. The company has already seen significant success, with revenue growing 8x since Q4 2024 and positive feedback from clients such as Peninsula Orthopaedic Associates. Assort Health is part of a growing trend of AI startups in healthcare, with companies like Abridge also making strides in the industry.Continuing with the latest tech news, NVIDIA faced a stock price drop after being banned from exporting chips to China, while China has also halted Boeing orders. The Fortune Most Powerful Women is hosting an international summit in Saudi Arabia where global economic forces will be discussed. Nobel prize winner Simon Johnson advised CEOs to make deals with the Trump administration to navigate through tariffs, suggesting investments in the US may be necessary.Shifting gears, the text discusses the development of Google's Dolphin Gemma AI model, which helps scientists study dolphin vocalizations. The model was created in partnership with Georgia Tech and the Wild Dolphin Project, using data collected since 1985 to train the AI. The use of AI in understanding animal communication showcases the potential of technology in bridging human understanding with animal communication.Moreover, Altman is planning to build a social media company to compete with another company, potentially for the purpose of gathering more data and pivoting OpenAI into surveillance. Elon Musk is already collecting a lot of data on residents, but instead of focusing on who will win this competition, it's important to compare the two companies.Lastly, the author argues that the concept of superintelligence is misunderstood by "AI doomers" who believe that advancing technology will lead to catastrophic consequences. Instead, they suggest that true intelligence lies in finding inner peace and emotional stability. Animals are often more at peace because they do not struggle with existential concerns like humans do. A truly superior being would prioritize achieving serenity and self-sustained bliss over technological progres

Thanks for listening! Follow us on Twitter, Instagram and Linkedin

Next Episode

2025-04-19 Saturday Special

Each Saturday we take the headlines of daily podcast and ask ChatGPT to create a Sci-Fi story inspired by the news. The outcome for this week is this short story.

Thanks for listening! Follow us on Twitter, Instagram and Linkedin