SP
BravenNow
Memories.ai is building the visual memory layer for wearables and robotics
| USA | technology | ✓ Verified - techcrunch.com

Memories.ai is building the visual memory layer for wearables and robotics

#Memories.ai #visual memory #wearables #robotics #artificial intelligence

📌 Key Takeaways

  • Memories.ai is developing a visual memory layer for wearables and robotics.
  • The technology aims to enhance devices' ability to record and recall visual information.
  • This innovation could improve user interaction and automation in wearable tech and robots.
  • The company focuses on creating a foundational layer for visual data processing.

📖 Full Retelling

Memories.ai is building a large visual memory model that can index and retrieve video-recorded memories for physical AI.

🏷️ Themes

Technology, Innovation

Entity Intersection Graph

No entity connections available yet for this article.

Deep Analysis

Why It Matters

This development matters because it represents a fundamental shift in how artificial intelligence systems interact with and learn from the physical world. It affects wearable technology users who could benefit from enhanced contextual awareness, robotics developers seeking more autonomous systems, and the broader AI industry pushing toward more human-like memory capabilities. The creation of a visual memory layer could enable devices to recognize patterns, recall past experiences, and make more informed decisions in real-time, potentially revolutionizing fields from healthcare monitoring to industrial automation.

Context & Background

  • Current AI systems typically process visual data in real-time without persistent memory, limiting their ability to learn from past experiences
  • Wearable technology has evolved from basic fitness trackers to sophisticated health monitors, but still lacks comprehensive contextual awareness
  • Robotics has advanced significantly in physical capabilities but remains limited in adaptive learning and environmental understanding
  • The concept of 'memory layers' in AI draws inspiration from human episodic memory systems that store and recall experiences
  • Previous attempts at visual memory in AI have focused on specific applications rather than creating a universal layer for multiple device types

What Happens Next

Memories.ai will likely release initial prototypes or partnerships with wearable manufacturers within 12-18 months, followed by integration into robotics platforms. Regulatory discussions about privacy and data ownership for visual memory systems will emerge as the technology develops. Expect competing platforms from major tech companies to enter this space within 2-3 years as the concept proves viable.

Frequently Asked Questions

What exactly is a 'visual memory layer'?

A visual memory layer is an AI system that continuously records, processes, and recalls visual information over time, allowing devices to build persistent understanding of their environment. Unlike current systems that process images in isolation, this creates connections between visual experiences to enable pattern recognition and contextual decision-making.

How will this technology protect user privacy?

Privacy protection will likely involve on-device processing where visual memories are stored locally rather than in the cloud, along with selective memory retention policies. The company will need to implement strong encryption and user controls over what visual information is stored and for how long, addressing concerns about constant visual recording.

What practical applications could this enable?

Applications could include wearables that remember where you left important items, health monitors that track symptom progression visually, and robots that learn optimal paths through environments. In healthcare, this could enable continuous visual monitoring of patient recovery, while in manufacturing, robots could remember complex assembly sequences.

How does this differ from existing computer vision technology?

Traditional computer vision analyzes single images or video frames in real-time without persistent memory. This technology adds temporal dimension by connecting visual experiences over time, enabling devices to recognize changes, learn from repetition, and make decisions based on historical visual context rather than just current input.

What are the biggest technical challenges facing this development?

Key challenges include managing massive amounts of visual data efficiently, developing algorithms that can identify meaningful patterns across time, and creating energy-efficient systems suitable for wearable devices. The technology must also solve the 'catastrophic forgetting' problem where AI systems lose old memories as they learn new information.

}
Original Source
Shawn Shen believes that AI will need to remember what it sees in order to succeed in the physical world. Shen’s company Memories.ai is using Nvidia AI tools to build the infrastructure for wearables and robotics to be able to remember and recall visual memories. Memories.ai announced a collaboration with semiconductor giant Nvidia at its GTC conference on Monday. Through this partnership, Memories.ai uses Nvidia’s Cosmos Reason 2, a reasoning vision language model, and Nvidia Metropolis, an application for video search and summarization, to continue to develop its visual memory technology. Shen (pictured above left) told TechCrunch that he and his co-founder and CTO Ben Zhou (pictured above right), got the idea for the company while building the AI system behind Meta’s RayBan glasses. Building the AI glasses got them thinking about how people would actually use the tech in real life if users couldn’t recall the video data they were recording. They looked around to see if they could find anyone already building that type of visual memory solution for AI. When they couldn’t, they decided to spin out of Meta and build it themselves. “AI is already doing really well in the digital world, what about the physical world?” Shen said. “AI wearables, robotics need memories as well. … Ultimately, you need AI to have visual memories. We believe in that future.” The ability for AI systems to remember, in general, is a relatively new. OpenAI updated ChatGPT to start to remember past chats in 2024 and fine-tuned that feature in 2025 . Elon Musk’s xAI and Google Gemini have launched their own memory tools too in the past two years. But these advancements have largely focused on text-based memory, Shen said. Text-based memory is much more structured and easier to index but aren’t as helpful for physical AI applications that largely interact with the world through sight and visuals. Techcrunch event Disrupt 2026: The tech ecosystem, all in one room Your next round. Your next hire. Y...
Read full article at source

Source

techcrunch.com

More from USA

News from Other Countries

🇬🇧 United Kingdom

🇺🇦 Ukraine