Hello Niuralogists!

‍

Welcome to this week's edition where we delve into the ever-evolving landscape of artificial intelligence. Our goal is to explore the latest breakthroughs and their profound implications across diverse domains, from workplaces and businesses to policies and personal experiences. In this issue, we'll uncover exciting developments, including Microsoft's latest AI Epoch and OpenAI's Strategic Partnership with Reddit.

‍

Stay informed with deeper insights...

‍

Microsoft's New AI Epoch

Microsoft has launched its new AI era at the Build Developer event, unveiling a range of new AI features and partnerships. Key announcements include the introduction of the Copilot+ Recall feature, which will operate locally, taking constant screenshots to remember user actions on Copilot+ PCs. Additionally, the company has released the new Phi-3-vision multimodal model and made the previously announced Phi-3-small and Phi-3-medium models generally available. Microsoft also introduced Team Copilot, an AI agent to assist users in Microsoft apps, and Copilot Studio, a tool for developers to create and deploy AI agents. Furthermore, a partnership with Khan Academy was announced, providing AI tools to all U.S. educators. These developments underscore Microsoft's ongoing efforts to challenge Apple's dominance in the market, particularly targeting the MacBook. The 'Recall' feature, with its potential for significant impact, requires substantial customer trust due to its always-watching nature.

‍

OpenAI Forms Strategic Partnership with Reddit

OpenAI has secured a significant partnership with Reddit, granting access to Reddit's real-time content through its data API. This collaboration enables OpenAI to integrate Reddit conversations into ChatGPT and other products, enhancing the precision and context-awareness of its AI models. This partnership mirrors a similar agreement Reddit had with Google, reportedly worth $60 million. Through this deal, OpenAI aims to improve its language models, making them more effective in processing human communication and natural language. Reddit will also benefit by developing new AI-powered tools for its users and moderators, potentially improving moderation and content understanding features. OpenAI will also serve as an advertising partner, helping Reddit offer more tailored ads. While this partnership promises significant advancements, it raises ethical concerns regarding user privacy and data usage. Reddit’s community, known for its vocal stance on executive decisions, may react cautiously to this development. This strategic move positions OpenAI to compete more directly with tech giants like Google and Microsoft, particularly in the social media space, while potentially reshaping Reddit's user experience and attracting more users.

‍

‍

Ilya Sutskever and Jan Leike Depart from OpenAI

OpenAI co-founder and chief scientist Ilya Sutskever has announced his departure from the company, following months of speculation stemming from the November 2023 ousting of Sam Altman. Sutskever expressed confidence in OpenAI's ability to build AGI that is both safe and beneficial under the current leadership. Alongside Sutskever, Jan Leike, co-lead of the superalignment group, also announced his resignation in a cryptic post. These departures come amid a wave of exits from OpenAI's superalignment and safety teams, fueling further speculation. In response, CEO Sam Altman has appointed Jakub Pachocki, a key researcher behind GPT-4, as the new chief scientist. While Sutskever’s departure resolves some of the tension, questions remain about the exodus from the safety team and the future destinations of these prominent AI figures.

‍

OpenAI's Sam Altman Declares AI Systems Like GPT-4 'Safe Enough for Use’

OpenAI CEO Sam Altman appeared at the Microsoft Build 2024 event, speaking with Microsoft CTO Kevin Scott about the future of AI systems like GPT-4, which he described as "safe enough for use." This appearance occurred shortly after actress Scarlett Johansson accused OpenAI of using a voice resembling hers in its GPT-4o chatbot. Altman did not address the controversy directly but emphasized his appreciation for the voice mode in GPT-4o and praised the rapid adoption and innovation driven by developers. He highlighted the significant progress from GPT-3 to GPT-4, noting improvements in intelligence, robustness, and safety. Altman described this period as a generational platform shift and urged developers to seize the moment. Despite internal challenges at OpenAI, including the departure of key executives and the dismantling of the superalignment team, Altman expressed confidence in the safety and utility of their AI models. He also mentioned new integrations with OpenAI announced at the conference, such as bringing GPT-4o to Copilot+ PCs, AI Azure Studio, and Microsoft Windows.

‍

‍

‍

Apple's iOS 18 AI with Eye-Tracking Technology

Apple has announced several new accessibility features for iOS 18, including AI-powered Eye Tracking, Music Haptics, and Vocal Shortcuts. Eye Tracking enables users to control their iPads or iPhones using just their eyes, without any additional hardware. Music Haptics allows users to feel music through the iPhone's Taptic Engine, and Vocal Shortcuts lets users create custom phrases for Siri to launch shortcuts and perform tasks. These features will be available later this year with updates like iOS 18 and iPadOS 18. While Eye Tracking may seem like a novelty, it represents a significant advancement in accessibility for users with physical disabilities. With various AI features and a rumored OpenAI partnership, Apple's WWDC conference on June 10th is highly anticipated by the AI community.

‍