AI Weekly: The AI Search Engine War and More – OpenAI, Google, Anthropic, and Meta Face Off

Welcome to another packed edition of AI Weekly! This week, AI giants went head-to-head with game-changing updates in search technology, apps, and even virtual assistants. OpenAI, Google, Anthropic, Meta, and others are redefining how we interact with AI, and their latest moves could change the AI landscape for good. Let’s dive into the highlights!

🚀 OpenAI: Search for All Pro Users and Chat History Upgrades

OpenAI is rolling out a new web search feature for all ChatGPT Pro users. This is similar to what users may have seen in AI apps like Perplexity, but it’s fully integrated into ChatGPT’s familiar interface. Now, users can get web-based answers directly within the app. Along with search, OpenAI has also added a handy chat history search feature, making it easier to find and revisit old conversations on specific topics.

The timing couldn’t be better, as OpenAI ramps up its desktop app presence. The advanced voice feature is now available on both Mac and Windows, bringing a more conversational AI experience right to your desktop. For fans who enjoy Perplexity’s detailed responses, ChatGPT’s new search may seem a bit brief for now, but OpenAI’s updates are only just beginning.

🔍 Google’s AI Overview Expands and Gemini API Launches Search for Developers

Google’s AI-powered search overviews, previously exclusive to the U.S., have gone global, covering over 100 countries. When you ask Google a question, the results now include a brief AI summary along with traditional links—a feature that’s now part of everyday search worldwide. Plus, Google’s Gemini API now allows developers to embed Google’s web-based AI search directly into their apps, opening up even more possibilities for applications built on Google’s AI ecosystem.

In addition to AI-powered search, Google’s Gemini API enables AI integration in diverse products and services. Developers can use it for generating insights, providing real-time answers, and even offering AI-generated guidance across different industries.

🎙️ Anthropic Adds Voice Dictation and Expands on GitHub

Anthropic, OpenAI’s fast-growing competitor, has introduced voice dictation in its Claude app. While it doesn’t offer audio responses like ChatGPT, the dictation feature improves accessibility and user interaction. In a fascinating twist, Anthropic also landed on GitHub Copilot, Microsoft’s AI-powered coding assistant, allowing users to choose between Claude, Google’s Gemini, or OpenAI’s models within GitHub. This expansion could shake up GitHub’s partnership landscape as Microsoft, an OpenAI investor, now broadens its horizons by integrating competing AI models.

🛠️ Meta’s Rumored Search Engine and Open-Source Podcast Generator

Meta is reportedly developing its own AI-powered search engine to rival OpenAI, Google, and other industry players. This follows its recent partnership with Reuters, hinting that Meta aims to integrate up-to-date, reputable sources into its search platform.

On the creative front, Meta has also released an open-source version of Google’s podcast generator, letting developers build and customize AI-powered podcast hosts. This offering aligns with Meta’s open-source approach, inviting innovation and experimentation from the community.

📱 Apple’s iOS 18.1 Rolls Out and New Mac Hardware Showcases M4 Chips

Apple users are getting a boost in AI features with iOS 18.1, which introduces tools like refined writing, notification summarization, and image editing. While this new OS version supports advanced AI functions, it’s currently limited to the latest iPhone models, like the iPhone 15 Pro. Alongside this, Apple’s new Mac lineup includes the M4 chip, explicitly built to support Apple Intelligence for AI-driven processing power.

These new devices bring Apple closer to integrating AI seamlessly into everyday tasks. However, the positioning of the power button on the Mac Mini—a minor but notable design choice—has users scratching their heads, as it requires lifting the device to turn it on.

🖌️ Recraft’s New AI Image Generator Tops Benchmarks

Recraft, a new player in image generation, has unveiled its V3 model and quickly topped aesthetic leaderboards with a remarkable 72% win rate against established models like MidJourney and Ideogram. Recraft’s standout feature is its ability to generate images with long text prompts, offering a more flexible tool for content creators who need high precision in their visuals.

🤖 Boston Dynamics and Physical Intelligence Bring Robots to New Heights

In robotics, Boston Dynamics’ autonomous sorting robot showcases the future of self-sufficient robots capable of independently managing logistics tasks. On the AI robotics side, Physical Intelligence’s Pi-Z model presents a revolutionary design where robots learn physical tasks by observation. From folding laundry to setting tables, these models could transform industries that rely on repetitive manual tasks.

The world of AI isn’t slowing down. From AI-powered search wars to the latest robotics and creative tools, each advancement marks a step toward a future where AI is deeply integrated into daily life. Stay tuned to AI Weekly for next week’s highlights—because, as we’ve seen, AI waits for no one.