AI Agents
/agents261
Discuss all AI agents, the technology that goes into them, and the implications because of them.
During a recent fireside chat, Sam Altman said that their latest AI model, "o1," has reached an advanced stage, which he referred to as level 2.
He hinted that the next big leap, level 3, will involve AI that can act on its own, like fully autonomous agents that can make decisions and carry out tasks without human guidance, and this might be coming sooner than we think.
OpenAI's 5 levels of Artificial Intelligence:
• Level 1: Chatbots, AI with conversation language
• Level 2: Reasoners, human-level problem solving
• Level 3: Agents, systems that can take actions
• Level 4: Innovators, AI that can aid in invention
• Level 5: Organizations, AI that can do the work of an organization
https://x.com/slow_developer/status/1836647241358143534?t=IPtA18RfrnTgIAG8snLu7w&s=08
He hinted that the next big leap, level 3, will involve AI that can act on its own, like fully autonomous agents that can make decisions and carry out tasks without human guidance, and this might be coming sooner than we think.
OpenAI's 5 levels of Artificial Intelligence:
• Level 1: Chatbots, AI with conversation language
• Level 2: Reasoners, human-level problem solving
• Level 3: Agents, systems that can take actions
• Level 4: Innovators, AI that can aid in invention
• Level 5: Organizations, AI that can do the work of an organization
https://x.com/slow_developer/status/1836647241358143534?t=IPtA18RfrnTgIAG8snLu7w&s=08
Been diving deep into the latest on AI agents, and these are three resources that helped me understand where we’re at and where we’re headed:
1. open.spotify.com/episode/786I...
- agent-based systems and current applications.
2.arxiv.org/html/2404.11...
- A solid read on multi-agent architectures.
3. www.youtube.com/watch?v=sal7.. .breaking down where we are in the AI agent cycle.
Perfect for anyone trying to keep up
1. open.spotify.com/episode/786I...
- agent-based systems and current applications.
2.arxiv.org/html/2404.11...
- A solid read on multi-agent architectures.
3. www.youtube.com/watch?v=sal7.. .breaking down where we are in the AI agent cycle.
Perfect for anyone trying to keep up
I often get asked: "Have you ever used OLAS (Agents)?"
Ofc I do, every day: I run Pearl
Pearl positions Agentic AI as potentially the most powerful UI/UX portal to Web3
All it takes to run an OLAS Agent is to download Pearl (on Mac or Windows), fund your Gnosis Chain MultiSig with 40 OLAS and 11.5 xDAI and start the agent
The Agent needs to run every day for 30 mins or else gets evicted
https://olas.network/operate
Ofc I do, every day: I run Pearl
Pearl positions Agentic AI as potentially the most powerful UI/UX portal to Web3
All it takes to run an OLAS Agent is to download Pearl (on Mac or Windows), fund your Gnosis Chain MultiSig with 40 OLAS and 11.5 xDAI and start the agent
The Agent needs to run every day for 30 mins or else gets evicted
https://olas.network/operate
Found this very interesting write-up about Agentic Protocols: https://goodalexander.com/posts/agentic-protocols/
Includes a number of interesting theses that all conclude how AI Agents will prevail 🤖 🔥
Includes a number of interesting theses that all conclude how AI Agents will prevail 🤖 🔥
Posting this to everyone with an interest in Agentic AI 🤖
190996
nonstopTheo🎩
@nonstoptheo·11:15 17/06/2024
New article out by Valory - the core contributor to the OLAS protocol (@olassians) - about Synthetic Data x OLAS agents
TL;DR:
The article discusses the importance of synthetic data in training AI models, emphasizing its role in preserving privacy and overcoming data scarcity.
It highlights the concept of agentic AI, which involves autonomous AI agents operating in decentralized environments, leveraging blockchain for transparency and fairness.
Olas ecosystem aims to democratize AI by enabling co-ownership, ensuring the benefits are widely shared and risks are minimized.
The convergence of advancements in web technology, AI, and user experience is leading to a new paradigm of autonomous, decentralized AI systems that are accessible and equitable.
https://www.valory.xyz/post/agent-economies-synthetic-data
TL;DR:
The article discusses the importance of synthetic data in training AI models, emphasizing its role in preserving privacy and overcoming data scarcity.
It highlights the concept of agentic AI, which involves autonomous AI agents operating in decentralized environments, leveraging blockchain for transparency and fairness.
Olas ecosystem aims to democratize AI by enabling co-ownership, ensuring the benefits are widely shared and risks are minimized.
The convergence of advancements in web technology, AI, and user experience is leading to a new paradigm of autonomous, decentralized AI systems that are accessible and equitable.
https://www.valory.xyz/post/agent-economies-synthetic-data
Are people here following OLAS (aka @olassians) and their Agents?
Shipped a user friendly tool to run Prediction market trader agents
- Abstracts away all code complexities
- Rewards $OLAS
- Only requires 20 $OLAS + 10 xDAI (runs on Gnosis) to get started
Happy to share more insights. Gauging interest
Shipped a user friendly tool to run Prediction market trader agents
- Abstracts away all code complexities
- Rewards $OLAS
- Only requires 20 $OLAS + 10 xDAI (runs on Gnosis) to get started
Happy to share more insights. Gauging interest
8531
Dean Pierce 👨💻🌎🌍
@deanpierce.eth·15:59 13/03/2024
GM, time for DeepMind to drop another revolutionary breakthough ♥️
https://deepmind.google/discover/blog/sima-generalist-ai-agent-for-3d-virtual-environments/
https://deepmind.google/discover/blog/sima-generalist-ai-agent-for-3d-virtual-environments/
9991
itarutomy.eth
@itarutomy.eth·23:49 23/02/2024
The missing pieces of AI agents: vast goals, state buildup, high skill in risky tasks, missing features, autonomous goals, and hidden operation.
https://www.lesswrong.com/posts/6BB9yqXKYjwnajYpM/does-increasing-the-power-of-a-multimodal-llm-get-you-an
https://www.lesswrong.com/posts/6BB9yqXKYjwnajYpM/does-increasing-the-power-of-a-multimodal-llm-get-you-an
295084
Polywrap
@polywrap·19:34 01/03/2024
231923
Health XChange
@healthxchange·17:56 21/02/2024
One of our biggest predictions I’d that in the next 5-10 years, everyday people will get access to a trained AI agent that will help them navigate their health
Ie. It does the same or better job as status quo (an overworked doc coming in and talking to you for 10-15 minutes)
Ie. It does the same or better job as status quo (an overworked doc coming in and talking to you for 10-15 minutes)
13901
Marissa
@marissaposner·02:01 15/02/2024
"We show that LLM agents can autonomously hack websites, performing tasks as complex as blind database schema extraction and SQL injections without human feedback. Importantly, the agent does not need to know the vulnerability beforehand." 😅 https://t.co/Bkc2gupDWL
283585
Alberto Enrique Duhau
@wildbuffalo·12:19 16/02/2024
I’ll start: AI Agent crypto enabling systems. Why? AI agents are the perfect use case for crypto given the digital nature of these agents, and if you consider the immense value they may create for humanity in the future they will need rails to transfer value easily, there will be no higher value use case in crypto.
1408
Mason Nystrom
@masonnystrom·17:15 12/12/2023
Lots of interesting early experiments of AI agents in crypto including:
- NPC consumer app agents (frenrug)
- Smart wallets with AI agents (Dawn)
- Gaming agents like Parallel's Colony game
- AI agent marketplaces
What other types of AI agents are people excited about?
https://nystrom.substack.com/p/crypto-ai-agents
- NPC consumer app agents (frenrug)
- Smart wallets with AI agents (Dawn)
- Gaming agents like Parallel's Colony game
- AI agent marketplaces
What other types of AI agents are people excited about?
https://nystrom.substack.com/p/crypto-ai-agents
10111
Bayram
@bayka.eth·05:42 16/02/2024
Coded an agent that scans my YouTube subscriptions for new videos, thanks them by relatedness to the product I am building right now, and suggests ideas from those videos
1171
𝚐𝔪𝟾𝚡𝚡𝟾
@gm8xx8·03:00 09/02/2024
More Agents Is All You Need
☺︎
⇩
finds the performance of llms improves w/ more agents using a simple sampling-and-voting technique. this approach, effective across different tasks, adds to existing methods for enhancing llms. (note: its effectiveness varies w/task difficulty)
https://arxiv.org/abs/2402.05120
☺︎
⇩
finds the performance of llms improves w/ more agents using a simple sampling-and-voting technique. this approach, effective across different tasks, adds to existing methods for enhancing llms. (note: its effectiveness varies w/task difficulty)
https://arxiv.org/abs/2402.05120
5525
jake
@jl·11:34 08/02/2024
“We introduce SELF-DISCOVER, a general framework for LLMs to self-discover the task-intrinsic reasoning structures to tackle complex reasoning problems that are challenging for typical prompting methods.”
https://arxiv.org/abs/2402.03620
https://arxiv.org/abs/2402.03620
1171
𝚐𝔪𝟾𝚡𝚡𝟾
@gm8xx8·01:53 07/02/2024
AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls
⇩
https://arxiv.org/abs/2402.04253
⇩
https://arxiv.org/abs/2402.04253
1171
𝚐𝔪𝟾𝚡𝚡𝟾
@gm8xx8·01:53 07/02/2024
AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls
⇩
https://arxiv.org/abs/2402.04253
⇩
https://arxiv.org/abs/2402.04253
1171
𝚐𝔪𝟾𝚡𝚡𝟾
@gm8xx8·03:36 06/02/2024
Large Language Model based Multi-Agents: A Survey of Progress and Challenges
↓
https://arxiv.org/abs/2402.01680
↓
https://arxiv.org/abs/2402.01680
1171
𝚐𝔪𝟾𝚡𝚡𝟾
@gm8xx8·05:41 01/02/2024
Efficient Tool Use with Chain-of-Abstraction Reasoning
CoA enhances LLMs w/ abstract reasoning training, leading to a 6% accuracy boost and 1.4x faster inference.
↓
https://arxiv.org/abs/2401.17464
CoA enhances LLMs w/ abstract reasoning training, leading to a 6% accuracy boost and 1.4x faster inference.
↓
https://arxiv.org/abs/2401.17464
