Chinese researchers unveil MemOS, the first ‘memory operating system’ that gives AI human-like recall
https://venturebeat.com/ai/chinese-researchers-unveil-memos-the-first-memory-operating-system-that-gives-ai-human-like-recall/
Chinese researchers unveil MemOS, the first ‘memory operating system’ that gives AI human-like recall
https://venturebeat.com/ai/chinese-researchers-unveil-memos-the-first-memory-operating-system-that-gives-ai-human-like-recall/
4 Comments
“Current AI systems face what researchers call the “memory silo” problem — a fundamental architectural limitation that prevents them from maintaining coherent, long-term relationships with users.
The system, called MemOS, treats memory as a core computational resource that can be scheduled, shared and evolved over time — similar to how traditional operating systems manage CPU and storage resources. The research demonstrates significant performance improvements over existing approaches, including a 159% boost in temporal reasoning tasks compared to OpenAI’s memory systems.
MemOS could represent a significant advancement in building AI systems that maintain context and improve over time, rather than treating each interaction as isolated.”
Idk man, kinda scared but kinda stoked? Like yeah, AIs with human-like recall probs gonna revolutionize loads of stuff but also feels like we’re inchin’ a bit closer to Skynet, ya know? Never thought I’d see the day we’re out here makin’ our own downfall lol.
Memory is def very important. Necessary if we ever want something close to AGI since currently the AI’s are basically wake up, get instructions follow them then fall asleep forgetting everything they just did never learning anything. If a human was like this they would be a vegetable as the vast majority of our intelligence is built up and not directly drawn from instinctual information we are born with.
I think giving AI long-term memory is essential for building AGI. But at the same time, I believe users will always want the ability to reset that memory—for privacy or emotional reasons.
If that becomes the norm, we might end up with AIs that seem to remember, but are ultimately designed to forget. That raises questions about whether such memory can truly support growth, relationships, or responsibility.