Daily AI Newsletter: Sam Altman: "We are gonna steamroll you" 🚂

In today’s email:

🍎 Apple's iOS 18 AI will be on-device preserving privacy, and not server-side
🧮 State tax officials are using AI to go after wealthy payers
📚 How to Fine-tune GPT3.5 Turbo for Custom Use Cases
🧰 8 new AI-powered tools and resources. Make sure to check the online version for the full list of tools.

Sam Altman: "We are gonna steamroll you" 🚂

The dialogue revolves around two predominant strategies in building AI technologies as discussed by Sam Altman, CEO of OpenAI, and Brad Lightcap. The first strategy focuses on assuming AI models will not improve significantly, leading companies to stack additional features onto existing models. The second strategy, which Altman advocates, is to anticipate continuous and significant improvements in AI models, aligning with OpenAI's progressive development trajectory. Despite the logical appeal of the latter, many startups still adopt the former approach, potentially limiting their future growth and adaptability.

Altman and Lightcap discuss the inherent challenges and convictions that come with advancing AI technology. They express a firm belief in the potential of deep learning and the advantages of scalability that have been evident since OpenAI’s inception. This conviction, coupled with continuous empirical support from ongoing projects, fuels their dedication despite external skepticism. Their discussion also touches on the cultural and operational dynamics within OpenAI, emphasizing the unique blend of research-driven innovation and pragmatic business applications, which are crucial as the organization scales.

The interview illuminates the broader impact and applications of AI, from enterprise adoption to societal benefits like healthcare. It also candidly addresses the challenges of rapid growth, such as maintaining a mission-driven approach amidst hype and external pressures. Both Altman and Lightcap reflect on personal lessons learned and the evolving landscape of AI adoption, stressing the importance of flexible, forward-thinking strategies that accommodate the fast pace of AI development and its transformative potential across various sectors.

Data Power-Up with Bright Data

Bright Data elevates businesses by collecting web data into turning it into actionable insights. Our global proxy network and web unblocking tools enable businesses to build datasets in real time and at scale. They provide a competitive edge in Ecommerce, Travel, Finance, and beyond.

Tap into the value of clean and structured data for market research, ML/AI development, and strategic decision-making. With our scalable solutions, you can efficiently enhance your data strategy, ensuring you're always one step ahead. Start with our free trial and see the difference yourself.

Apple's iOS 18 AI will be on-device preserving privacy, and not server-side

Apple's upcoming iOS 18 is set to enhance Siri and other features with advanced AI capabilities that prioritize user privacy by processing data directly on devices, rather than relying on cloud services. Bloomberg's Mark Gurman has reported that the initial AI features in iOS 18 will operate entirely on-device, allowing them to function without an internet connection. This shift is part of Apple's broader strategy to integrate more privacy-focused AI tools into its operating systems, including the potential use of an in-house developed large language model known as "Ajax."

Despite these advancements, more complex AI functionalities will still require internet access. AppleInsider notes that while basic text analysis and response generation will be available offline, more sophisticated features will not. The move to on-device processing is aimed at mitigating privacy concerns and the issue of AI "hallucination," where models generate confident but inaccurate or fabricated information. This change could enhance the reliability of AI applications in environments where data security is paramount.

Looking forward, Apple plans to unveil more about its AI strategies at the Worldwide Developers Conference (WWDC) starting on June 10. The development and incorporation of on-device AI could position Apple as a leader in secure and private AI solutions, distinguishing its products from competitors that rely on cloud-based AI technologies. This strategic focus aligns with growing consumer demand for more secure and trustworthy AI applications in everyday technology.

Amazon Music follows Spotify with an AI playlist generator of its own, Maestro

Amazon Music has introduced a new feature called Maestro, an AI-driven playlist generator, currently in beta testing in the U.S. for iOS and Android users. This tool allows users to create customized playlists by entering prompts, which can include text, emojis, or a combination of both. Suggestions are available for those unsure what to write, and the AI then generates a playlist that aims to reflect the themes of the prompts. This service is available to a select group of Amazon Music's free users, Prime members, and Unlimited subscribers.

Maestro is designed with certain limitations and protections; for instance, it includes mechanisms to block offensive language and inappropriate content. While the service is in its early stages and may not always deliver perfect results, it represents Amazon's attempt to compete with Spotify, which has also recently tested a similar AI feature in the U.K. and Australia. Amazon's approach differentiates by providing immediate playlist access and save features for subscribers, whereas free and Prime users can only preview 30-second clips of the songs.

To use Maestro, users must update to the latest version of the Amazon Music app and select the Maestro option from the home screen or when creating a new playlist. The feature supports both spoken and written prompts, with playful examples provided like "😭 and eating 🍝" or "Music my grandparents made out to." Amazon plans to expand access to Maestro gradually, promising more users will experience its AI playlist capabilities in the future.