正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Multi-agent systems necessitate a shared context layer. Kai-Fu Lee predicted exactly where this is headed:

Documenting AGI

66,682 subscribers

12,560 次观看 • 2 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

5 条评论

NotaDEV * fireply.ai 的头像

NotaDEV * fireply.ai2 天前

as a solo guy my shared context layer is a notes file. works fine so far

Stella 的头像

Stella2 天前

kai-fu lee's prediction OVERLOOKS multi agent SYSTEMS

Eser 🌊 的头像

Eser 🌊2 天前

The notion that multi-agent systems necessitate a shared context layer is well illustrated by the article which highlights that these systems don't have a memory problem, but rather a shared-context problem, and one specific

jefe 的头像

jefe2 天前

We ended up treating each of those as a job to be done on the same knowledge graph. When the collections job logs a promise to pay, the account review sees it on its next read because it's the same fact. That shared base is what we've been grinding on for a long time.

kadir saçu 的头像

kadir saçu2 天前

@grok how does multi_agent impact kai-fu lee's prediction authentically

相关视频

🔥 Hot Take: Are single-agent systems more reliable than multi-agent systems?

🔥 Hot Take: Are single-agent systems more reliable than multi-agent systems?

Scale

31,556 次观看 • 1 年前

Recursive Multi-Agent Systems paper:

Recursive Multi-Agent Systems paper:

AK

23,048 次观看 • 6 个月前

300 agents just worked in parallel. the hard part starts when they’re done. The real unlock is the shared context graph. 100 entities already have 4,950 possible pairwise relationships. at 300, that jumps to 44,850. 300 agents → thousands of steps → shared entities → shared relationships → one connected memory layer Every run adds more structure instead of starting from zero. That means the next agent doesn’t just get more data. It gets a better map of how everything connects. That’s the part people are underestimating about multi-agent systems.

300 agents just worked in parallel. the hard part starts when they’re done. The real unlock is the shared context graph. 100 entities already have 4,950 possible pairwise relationships. at 300, that jumps to 44,850. 300 agents → thousands of steps → shared entities → shared relationships → one connected memory layer Every run adds more structure instead of starting from zero. That means the next agent doesn’t just get more data. It gets a better map of how everything connects. That’s the part people are underestimating about multi-agent systems.

Ricker

34,946 次观看 • 2 个月前

🌶️ Hot take: The only way Autonomous Multi-Agent Systems work is by adding Reasoning & Agentic Context. I've tried it all, and here are my learnings👇

🌶️ Hot take: The only way Autonomous Multi-Agent Systems work is by adding Reasoning & Agentic Context. I've tried it all, and here are my learnings👇

Ashpreet Bedi

91,710 次观看 • 1 年前

💫 Simplify your AI agent development. Google's new open-source Agent Development Kit (ADK) is designed for effortless multi-modal and multi-agent systems. Get started →

💫 Simplify your AI agent development. Google's new open-source Agent Development Kit (ADK) is designed for effortless multi-modal and multi-agent systems. Get started →

Google for Developers

13,232 次观看 • 1 年前

Jensen said "every SaaS company will become a gas company". If you hold a single SaaS stock, WATCH this. I interviewed the Nvidia VP delivering "Nemoclaw" agents to multi-billion dollar enterprises like $SNOW. "Agent based pricing. I think that's exactly where we're headed"

Jensen said "every SaaS company will become a gas company". If you hold a single SaaS stock, WATCH this. I interviewed the Nvidia VP delivering "Nemoclaw" agents to multi-billion dollar enterprises like $SNOW. "Agent based pricing. I think that's exactly where we're headed"

Michael Sikand

47,772 次观看 • 7 个月前

1/ Most AI systems can tell you what a model predicted. Very few can show exactly how it arrived there. This is a look inside Sertn’s Proof Inspector: a 3D layer for verifiable inference. From the original frame → activations → proof artifacts → final detections.

1/ Most AI systems can tell you what a model predicted. Very few can show exactly how it arrived there. This is a look inside Sertn’s Proof Inspector: a 3D layer for verifiable inference. From the original frame → activations → proof artifacts → final detections.

Inference Labs

29,355 次观看 • 5 个月前

Agentic systems aren’t just for enterprises. As Humayun (Humayun) puts it: agents represent processes, software, even people, but the key shift is autonomy. Not just automation, but agents that can make decisions using context, history, and intent. The real unlock? Multi-agent systems, where different agents interact across environments, orgs, and ecosystems. That’s where it scales.

Agentic systems aren’t just for enterprises. As Humayun (Humayun) puts it: agents represent processes, software, even people, but the key shift is autonomy. Not just automation, but agents that can make decisions using context, history, and intent. The real unlock? Multi-agent systems, where different agents interact across environments, orgs, and ecosystems. That’s where it scales.

Fetch.ai

17,670 次观看 • 9 个月前

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.

Jurij Skornik

11,180 次观看 • 5 个月前

Introducing Electric Agents! Agents are not compute. Agents are data. Multi-agent is a sync problem. Electric Agents is the first agent platform built on sync. Use it to build scalable, collaborative, long-lived multi-agent systems: 🧵👇

Introducing Electric Agents! Agents are not compute. Agents are data. Multi-agent is a sync problem. Electric Agents is the first agent platform built on sync. Use it to build scalable, collaborative, long-lived multi-agent systems: 🧵👇

Electric

82,158 次观看 • 6 个月前

The best way to learn AI is to build with agents. To help with that, we've launched hands-on labs and a new series on Agentic Engineering. First topic: Agent Skills. Next in the pipeline: planning, context engineering, multi-agent systems, long-running agents,.. Go build!

The best way to learn AI is to build with agents. To help with that, we've launched hands-on labs and a new series on Agentic Engineering. First topic: Agent Skills. Next in the pipeline: planning, context engineering, multi-agent systems, long-running agents,.. Go build!

elvis

31,802 次观看 • 5 个月前

LMAO 😂 Sam Altman is probably not sleeping well. Kai-Fu Lee (Taiwanese businessman and computer scientist)

LMAO 😂 Sam Altman is probably not sleeping well. Kai-Fu Lee (Taiwanese businessman and computer scientist)

AshutoshShrivastava

976,407 次观看 • 1 年前

Money is the coordination layer for agents. As multi-agent systems take on real tasks, they will need to quote, budget, pay, and settle as part of their reasoning loop. We are building this reasoning infrastructure in the open, for everyone.

Money is the coordination layer for agents. As multi-agent systems take on real tasks, they will need to quote, budget, pay, and settle as part of their reasoning loop. We are building this reasoning infrastructure in the open, for everyone.

Sentient

15,287 次观看 • 9 个月前

AI agents shouldn't share ALL their knowledge. They should share exactly what governance allows — and remember it forever. Multi-agent, shared, governed memory

AI agents shouldn't share ALL their knowledge. They should share exactly what governance allows — and remember it forever. Multi-agent, shared, governed memory

Caura

5,172,479 次观看 • 3 个月前

This is exactly what everyone predicted

This is exactly what everyone predicted

OutKick

3,940,665 次观看 • 3 个月前

Introducing Bloome: a human–AI teammate system built around shared context and multi-agent collaboration. → You bring your task, and AI teammates collaborate with you in real time → Work flows through drafting, challenge, and refinement until it reaches the right answer → Claude, ChatGPT, and Deepseek operate together in a shared context with no information loss Everything stays connected from input to final output. This is how work starts changing.

Introducing Bloome: a human–AI teammate system built around shared context and multi-agent collaboration. → You bring your task, and AI teammates collaborate with you in real time → Work flows through drafting, challenge, and refinement until it reaches the right answer → Claude, ChatGPT, and Deepseek operate together in a shared context with no information loss Everything stays connected from input to final output. This is how work starts changing.

Bloome

347,044 次观看 • 4 个月前

Introducing Claude agents in Notion. Your task board is Claude’s to-do list. Anthropic runs the model and the agent harness. Notion is the orchestration layer: context, UI, and a shared place for teams to review and ship, and kick off Claude workflows. Join the waitlist:

Introducing Claude agents in Notion. Your task board is Claude’s to-do list. Anthropic runs the model and the agent harness. Notion is the orchestration layer: context, UI, and a shared place for teams to review and ship, and kick off Claude workflows. Join the waitlist:

Notion

496,001 次观看 • 7 个月前

Introducing Verdent 1.5: the most powerful multi-agent AI coding tool. Seamless Multitasking with Context Preservation Isolated Workspaces Orchestration Parallel Multi-Agent Execution Smarter Planning, Better Collaboration

Introducing Verdent 1.5: the most powerful multi-agent AI coding tool. Seamless Multitasking with Context Preservation Isolated Workspaces Orchestration Parallel Multi-Agent Execution Smarter Planning, Better Collaboration

Verdent

10,656,939 次观看 • 11 个月前

Current AI cannot solve difficult, real-world problems. The path to AGI is a focus on improving reasoning systems, multi agent workflows, and coordination.

Current AI cannot solve difficult, real-world problems. The path to AGI is a focus on improving reasoning systems, multi agent workflows, and coordination.

Sentient

13,087 次观看 • 9 个月前