Welcome to this week’s edition of the TekTonic newsletter, where we review the latest and most impactful AI and tech news for the first week of July 2024. Amidst growing skepticism towards OpenAI, Salesforce’s “Tiny Giant” model shows that innovation isn’t always about the size of models. The cybersecurity news landscape is full of eye-opening revelations, as evidenced by new vulnerabilities in AI models. Meanwhile, advancements in robotics, educational technology, and finance signal transformative shifts across various sectors. Join us as we explore these stories and more, reflecting today’s rapid evolution and broad impact of AI and technology.
My favorite story this week is essentially a theme. This week, an influential investor, Ari Emanuel, called Sam Altman, CEO of OpenAI, a con man who can’t be trusted. A sentiment that seems to be growing. Just as we heard a new report that OpenAI is silencing insiders, we heard about the potential inclusion of Google’s AI and possibly Anthropic’s AI in the next iPhone release. We also heard Apple appointed an observer to OpenAI’s board (no reason given). We also head Anthropic is paying other companies to test the safety of their systems as well as funding the creation of new trustworthy AI benchmarks. Is this an attempt to capitalize on OpenAI’s growing negative sentiment?
This week, we learned that OpenAI has responded to security concerns by encrypting chats on its ChatGPT desktop app for Mac, ensuring user privacy and reinforcing its commitment to data security. This move comes amidst an intriguing partnership announcement between Apple and Google. Meanwhile, Salesforce has introduced the “Tiny Giant” AI model, xLAM-1B, which outperforms larger counterparts for function calling. This demonstrates that innovation in AI isn’t always about size but efficiency and application.
Apple executive Phil Schiller has joined OpenAI as a board observer, highlighting the dynamic nature of corporate alliances. This collaboration between Apple and OpenAI, including Apple as an observer on the board, underscores the complex interplay between OpenAI and their shared interest in advancing AI technologies and possible distrust between the wider public and OpenAI.
Meta has released multi-token prediction models on the innovation front, promising to revolutionize language processing with a more nuanced understanding of language structure and context. Similarly, French AI Lab Kyutai released the Moshi model, an open-source multimodal AI. The key is that this multimodal model is capable of vision and speech out of the gates and not added later.
In cybersecurity, AI models’ vulnerabilities have come to the forefront. OpenAI’s GPT-4 has been found capable of generating a type of synthetic cancer, just as OpenAI experienced a significant hack and didn’t inform the FBI. This discovery and concerns over potential Skeleton Key attacks, as warned by Microsoft, underscore the need for robust security measures in AI development.
The week’s stories also shed light on the personal dynamics within the tech industry. Elon Musk’s announcement of the upcoming Grok 2 and Grok 3 releases and Sam Altman’s ambitious projections for GPT-5 indicate a future brimming with AI advancements that promise to push the boundaries of what is possible.
In other AI news, Microsoft introduced Copilot for Finance, a tool designed for human finance users but also capable of automating much of their work through AI and machine learning. This marks the fast-approaching moment when AI stops being a tool and becomes a synthetic employee. This milestone is largely going unnoticed by most media outlets.
Google introduced new Workspace AI tools aimed at empowering educators and students. These tools inspire creativity and enhance grading and collaboration, potentially transforming classroom dynamics. Similarly, Squirrel AI has debuted its enhanced large multimodal adaptive model, revolutionizing educational software with its advanced capabilities.
China has emerged as the leader in generative AI patent applications in politics and government, with the United States trailing behind. This revelation from the UN underscores the competitive landscape in AI innovation and its implications for global technological dominance. Meanwhile, the U.S. Army has issued new guidelines for generative AI and large language models, highlighting the military’s increasing reliance on advanced technologies.
In the business world, a recent survey reveals that 60% of companies prioritize generative AI over the next two years, contrasting conversations about plateauing AI progress and signaling a significant shift towards AI-driven solutions in various industries. Atlas Invest has launched an AI-powered bridge financing platform, unlocking billions in real estate investment. At the same time, Rockwell Automation’s report shows manufacturers focus on AI to build resiliency and drive sustainable growth.
Robotics continues to capture headlines, with Canadian Sanctuary AI securing $140 million in funding to develop human-like robots, marking a significant milestone in robotics. However, not all robot news is positive; a South Korean administrative robot’s fall down a set of stairs has led to reports of the first robot suicide in the country, raising questions about the integration of robots into daily life. On a more positive note, a new robot capable of jumping higher than any other could one day be used to explore planets, showcasing the potential for robotics in space exploration.
The media and entertainment industry is also experiencing a technological transformation. ElevenLabs has secured rights to use AI for narrations by deceased celebrities, offering a unique twist to audiobook and media content. YouTube’s new policy allows users to request the removal of AI-generated deepfakes, addressing growing concerns over digital content manipulation. In a groundbreaking achievement, independent filmmaker Hooroo Jackson created “DreadClub: Vampire’s Verdict,” the world’s first fully AI-generated animated feature, setting a new standard for AI filmmaking.
SentiAR’s hands-free headset lets surgeons view 3D, mixed-reality, real-time images of patients’ hearts, revolutionizing cardiology and surgical precision. This technology represents a significant advancement in medical imaging and patient care.
This week, the startup scene has seen two announcements: Redactive AI has secured $11.5 million in seed funding to develop secure generative AI agents, and Magic, an AI coding startup, seeks a $1.5 billion valuation in a new funding round. These investments highlight the growing interest and potential in AI-driven software development.
Finally, in the crypto and blockchain world, a $200 million investment is being made in betting on integrating AI with blockchain technology, attracting attention from prominent investors like Peter Thiel.
That is it. Stay tuned for more news next week. I APPRECIATE ALL THE READERS. Help spread the word. I put these newsletters together so productive humans like you can stay on top of the latest AI stories shaping our world.
If you like these weekly tech news reports, subscribe to get notified of new editions and updates. For daily updates, you can check out our news page. Or, for a more in-depth analysis of the week’s news, sign up to receive our free weekly newsletter located to the right of the daily news or follow me on Twitter or YouTube.
