Daily AI Newsletter: GPT-4o - OpenAI's new flagship model

In today’s email:

🤑 AI startups in the Bay Area raised $27B+ in 2023, more than 50% of global AI funding
👀 Sam Altman did an AMA on Reddit
🤯 Creepy Study Suggests AI Is The Reason We've Never Found Aliens
🧰 9 new AI-powered tools and resources. Make sure to check the online version for the full list of tools.

OpenAI debuts GPT-4o ‘omni’ model now powering ChatGPT

Yesterday OpenAI unveiled their latest AI model, GPT-4o, during a presentation in San Francisco. The "o" in GPT-4o stands for "omni," highlighting the model's versatility across text, speech, and video. This iteration, according to OpenAI CTO Mira Murati, extends GPT-4’s functionalities, enabling it to manage tasks involving voice and vision in addition to text. GPT-4o is designed to enrich user interaction with machines, providing more intuitive and dynamic engagement capabilities.

GPT-4o enhances the functionalities of OpenAI’s ChatGPT by introducing upgraded voice and vision capabilities. Now, users can interact with ChatGPT in a conversational manner, interrupting the AI and receiving real-time responses. The model can also handle images and video inputs, allowing it to perform tasks like identifying objects in photos or translating text from images. This leap in capabilities is part of OpenAI's broader aim to make user experiences with AI more seamless and natural.

The increased speed of response, down to 320 milliseconds, enhances the natural flow of conversation, allowing for real-time interactions without awkward pauses. This improvement is evident in various demos, such as real-time translation and the AI's ability to interpret visual cues for tasks like solving math problems. The addition of non-verbal instructions, such as pointing out parts of a geometry problem with an Apple Pencil, showcases the potential for more intuitive and interactive learning experiences.

Moreover, OpenAI announced several updates to its platforms alongside the launch of GPT-4o. A refreshed user interface for ChatGPT and a new desktop app are aimed at improving accessibility and convenience. OpenAI also introduced a freemium model for the GPT Store, expanding access to its AI tools. These updates reflect OpenAI’s ongoing commitment to enhancing the intelligence and usability of its products, making advanced AI more accessible to a broader audience.

Related:
- GPT-4o has EQ now
- GPT-4o takes #1 & #2 on the Aider LLM leaderboards
- OpenAI’s custom GPT Store is now open to all for free
- Disarmingly lifelike: ChatGPT-4o will laugh at your jokes and your dumb hat

Web publishers brace for carnage as Google adds AI answers

Google's introduction of AI-generated answers in its search results is causing significant concern among web publishers and content creators. These AI responses, part of the Search Generative Experience (SGE), provide detailed answers to user queries, pushing traditional links further down the page. This change threatens the traffic and revenue of websites that rely heavily on Google for visibility. Experts predict that web traffic from search engines could drop by 25% by 2026, potentially causing billions in losses for creators and publishers.

Creators like Kimber Matherne, who runs the popular food blog Easy Family Recipes, are particularly worried about the impact on their businesses. Matherne's blog, which receives 40% of its traffic from Google, may see a significant decline in visitors due to AI answers summarizing her content without directing users to her site. Similarly, Jake Boly, a strength coach who reviews workout shoes, has already experienced a 96% drop in Google-driven traffic, making it harder for him to sustain his website.

The broader rollout of AI in Google Search, while aimed at providing quick and comprehensive answers, raises questions about the future of the open internet. Critics argue that it centralizes information control within a few large tech companies and diminishes the role of independent content creators. This shift not only threatens the livelihoods of many web publishers but also challenges the diversity and accessibility of information on the internet.

AI startups in the Bay Area raised $27B+ in 2023, more than 50% of global AI funding

The San Francisco Bay Area has solidified its position as the global leader in AI technology and funding, with over 50% of all global venture funding for AI-related startups in 2023 going to companies headquartered in the region. The trend began with OpenAI’s ChatGPT reaching 100 million users in early 2023 and further accelerated by OpenAI’s $10 billion funding from Microsoft. As a result, Bay Area AI startups raised over $27 billion in 2023, a significant increase from the $14 billion raised in 2022, making the region a magnet for AI talent and investment.

Prominent AI companies such as OpenAI, Anthropic, and Inflection AI have established significant real estate footprints in San Francisco, contributing to a resurgence in the city's commercial real estate market. The influx of AI-focused startups, many graduating from accelerators like Y Combinator and 500 Global, has led to increased demand for office space and created a vibrant ecosystem for networking, hiring, and innovation. Despite challenges in the broader tech sector, AI has emerged as a bright spot, attracting both domestic and international startups to the Bay Area.

While the Bay Area dominates AI funding, significant AI research and startup activity is also occurring globally, with notable contributions from universities and companies in Europe, Canada, and China. Nevertheless, the unique concentration of Big Tech, venture capital, and academic talent in the Bay Area, particularly from institutions like UC Berkeley and Stanford University, continues to drive its leadership in the AI sector. This dynamic environment has fostered a migration of talent back to the Bay Area, underscoring its status as the epicenter of the AI revolution.

Apple Nears Deal With OpenAI to Put ChatGPT on iPhone

Apple Inc. is close to finalizing an agreement with OpenAI to integrate ChatGPT technology into the upcoming iOS 18 operating system for iPhones. This move is part of Apple's broader strategy to enhance its devices with advanced artificial intelligence capabilities. Although discussions are also ongoing with Alphabet Inc.'s Google for its Gemini chatbot, these have not yet resulted in a deal.

The potential collaboration with OpenAI was highlighted as part of Apple's plans to introduce a range of new AI features at its annual Worldwide Developers Conference in June. These features are expected to run on data centers powered by Apple’s own in-house processors, highlighting the company's push into AI under the leadership of CEO Tim Cook, who has emphasized a thoughtful approach to integrating AI into Apple products.

During Apple’s recent earnings call, Tim Cook expressed his belief in the transformative potential of AI and pointed out Apple's unique position with its integrated ecosystem of hardware, software, and services. He suggested that these elements would give Apple a competitive advantage in the evolving AI landscape. Representatives from Apple, OpenAI, and Google have not provided comments on the ongoing discussions.

Other stuff

Internal Emails Reveal How a Controversial Gun-Detection AI System Found Its Way to NYC
IBM open-sources its Granite AI models - and they mean business
SoftBank's Arm plans to launch AI chips in 2025
Can China's AI Technology Compete With the US?
Google’s 3D video conferencing platform, Project Starline, is coming in 2025 with help from HP
Creepy Study Suggests AI Is The Reason We've Never Found Aliens
500+ AI engineers just spent 24 hours straight putting Llama3 to the test.
Stable Artisan: Media Generation and Editing on Discord
Sam Altman did an AMA on Reddit