Meta aims to develop an AI agent that will shop on Instagram on your behalf, performing the role of an agent.

      Meta's upcoming AI initiative is progressing from chatbots to a shopping assistant. An Engadget report, citing The Information, reveals that the company is developing a more sophisticated “agentic” AI assistant capable of performing tasks for users, rather than simply responding to requests. This includes a dedicated AI shopping assistant for Instagram.

      What's in the works at Meta?

      Bryan M. Wolfe / Digital Trends

      The new AI tool is reportedly based on a novel model called Muse Spark, which Meta is currently testing internally. Additionally, the company is working on another initiative, named Hatch, with testing expected to conclude by June. For those monitoring the AI assistant landscape, these tools are designed to do more than just summarize web pages or create text; they aim to navigate apps, engage with services, and manage multi-step tasks for users.

      Engadget labeled this project as a competitor resembling OpenClaw, especially considering how aggressively Meta is pursuing agentic AI. For reference, OpenClaw has rapidly gained recognition in this field, as it is intended to function across various software and hardware platforms. Hatch is reportedly being trialed on platforms like DoorDash, Reddit, and other third-party services.

      Despite this, Instagram likely serves as the simplest venue for Meta to publicly test this feature. The app already integrates creators, product discovery, advertising, recommendations, and shopping behaviors, making it ideal for an AI agent capable of searching for products, comparing choices, and assisting users with purchases.

      However, some concerns are evident. Meta has spent years converting user attention into advertisements, recommendations, and commerce. Offering an AI assistant from the same company may enhance convenience but could also raise unease.

      The Manus acquisition setback

      Recently, Meta tried to expedite its agent strategy by acquiring Manus, an AI agent startup that originated in China and later relocated to Singapore. The deal was valued at approximately $2 billion but was blocked by Chinese regulators in late April. Consequently, Meta's effort to launch a new AI agent initiative comes on the heels of the Manus deal's failure. With increased investment in AI infrastructure, it’s clear that the company desires a robust agent platform. If acquiring one proves difficult, building an in-house solution appears to be the logical alternative.

      ---

      Korea introduces a robotic Buddhist monk at a monastery, signaling a futuristic trend.

      Buddhism enters the age of robotics with a monk named Gabi.

      A humanoid robot participated in a Buddhist ceremony in Seoul, making for a fascinating event. The robot, named Gabi, was unveiled at Jogyesa Temple in central Seoul during a ceremony held in advance of Buddha’s Birthday celebrations. Standing 1.3 meters tall, the robot received the Dharma name "Gabi" during a special refuge ceremony led by the Jogye Order of Korean Buddhism, the largest Buddhist order in South Korea.

      Why a robot became a monk in an actual temple

      ---

      OpenAI's latest voice update makes Siri and Alexa appear outdated.

      The universal translator has transitioned from science fiction to an available app.

      OpenAI has released three new audio models within its Realtime API, which are significant for developers creating voice-activated applications. The models include GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper. Together, these advancements propel voice AI beyond basic dialogue into a domain that can comprehend user intent, take action, and maintain fluid conversations.

      ---

      ChatGPT now allows users to designate someone to check in during difficult times.

      OpenAI introduces a human safety feature in ChatGPT for crisis situations.

      AI chatbots have facilitated open discussions about various subjects, including some of the most challenging topics. However, this openness presents both benefits and drawbacks. OpenAI is taking steps to mitigate risks, introducing a new feature called Trusted Contact, which is beginning to appear in ChatGPT settings for adult users. This feature enables users to designate an individual who can be notified if ChatGPT identifies a significant self-harm issue.