# The Future of Work

Toran Bruce Richards, AutoGPT · Silen Naihin, AutoGPT · Reinier van der Leer, AutoGPT

AI Engineer Summit 2023

[Talk page](https://ai.engineer/talks/ZYoZSU58m_Y-future-work) · [Video](https://www.youtube.com/watch?v=ZYoZSU58m_Y)

## The Future of Work

AutoGPT’s roadmap connects conversational task requests to shared development tools, measurable progress, and the safety required for agents that act on a user’s behalf.

### Ask for the completed spreadsheet

[0:59](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=59s)

How do you fill a lead-generation spreadsheet when you start with company names and links? Silen Naihin’s opening example follows a familiar loop: search Google, copy a result, paste it into the sheet, visit LinkedIn, copy again, and return to Google. The work repeatedly crosses the same interfaces and websites to assemble the information one row at a time.

![A dense collage of keyboard keys labeled Copy and Paste, LinkedIn logos, and globes with search bars.](https://ai.engineer/api/talk-essay-media/7468df7cbb0f45cf27452bd634b230d956de81822f5b73a32885f1717e04ca80)

Copy-and-paste keys, LinkedIn logos, and web-search icons crowd the slide.

The proposed conversational alternative keeps the output unchanged: a completed spreadsheet of leads. Instead of directing every search and paste, the user asks for the result through chat. **The intended shift is from operating each interface to specifying the work that should be completed.**

The same proposal extends to two other workflows, in sequence:

- **Inbox work:** An overloaded inbox leads to hours spent sending variations of the same email. A conversational agent would handle that repetition, with the proposed business outcome of turning unread messages into leads.
- **Application development:** Silen invokes projects costing millions of dollars and taking weeks, months, or years, while developers still copy and paste from **ChatGPT**. Here, too, the proposed alternative is to request the work through chat.

These are motivating scenarios for conversational automation, not demonstrated implementations.

### From creative orchestration to a development community

[2:33](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=153s)

[**AutoGPT**](https://github.com/Significant-Gravitas/AutoGPT) gives these scenarios a broader ambition: agents execute routine digital work so that people can spend more of their effort directing creative work. In this vision, being an AI engineer becomes less about personally implementing every step and more about orchestrating what an agent should accomplish. It is a proposal for human augmentation, with more room for creative decisions and less repetitive execution.

Silen describes AutoGPT as the fastest repository to reach 100,000 GitHub stars. He connects that attention to widespread news coverage and growing interest in agents from major technology companies. Those observations concern enthusiasm for the approach; they do not establish how reliably an agent can complete the opening tasks.

The team then describes the community behind the repository. Its reported figures distinguish people registering interest from people contributing work and feedback:

| Participation | Reported scale at the talk | Contribution to the project |
|---|---|---|
| GitHub stars | 150,000, reached the week before the talk | An expression of interest |
| Contributors | More than 460 | Thousands of pull requests and issues |
| Online community | 47,000 members | Conversations and insights |

The distinction matters: stars record attention, while submitted changes, issue reports, and discussions give the project material to build with and learn from.

The team credits those contributions and conversations with what it had built and learned over its first six months. Its commitment is to keep growing and empowering that community, making participation part of how the future agent will be developed.

### Connect agent development through a shared protocol

[6:19](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=379s)

Work on the original open-source agent continues through capability improvements and the incorporation of research. Alongside that agent, the team introduces [**Forge**](https://github.com/Significant-Gravitas/AutoGPT), a standardized template intended to make it easier for other developers to build agents. The scope expands from developing one agent to supplying a starting point for other agent creators.

A development UI provides a way to interact with an agent and improve it iteratively. The displayed interface brings together connected task nodes, a `SortCsv` test card, and User and Agent panels displaying a JSON artifact. It makes the development workflow concrete: tasks, test controls, and an artifact can be inspected within the same interface.

![Connected task nodes appear beside a SortCsv test card and User and Agent panels displaying a JSON artifact.](https://ai.engineer/api/talk-essay-media/067e2f3928578e6b7404d43d682458cb56f4eb1d90e9de97efe57cb802d432f4)

The development interface combines a task graph, test controls, and an artifact panel.

These tools connect through [**Agent Protocol**](https://github.com/agi-inc/agent-protocol), which Silen attributes to the AI Engineer Foundation, alongside other industry standards. Its architectural role is a common interface between agents and the tools around them. An agent implementing that protocol can use the compatible benchmark, frontend development tool, and other offerings built on it. Compatibility depends on the shared contract rather than on every developer building the same agent.

The tools are still prerelease, but participants are already using them in a virtual hackathon with a reported $30,000 cash prize pool. That use supplies feedback, bug fixes, and insights for subsequent development. The hackathon therefore serves both as an invitation to build agents and as a source of practical experience with the tooling.

### Why prioritize coding agents?

[7:29](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=449s)

One lesson from this work is, in Silen’s words, “code is king.” The reasoning starts with the environment an agent must act within: code is the underlying fabric of digital systems. Coding ability is therefore positioned as a foundation for accomplishing many different digital tasks. The destination remains a generalist agent; the team’s strategic thesis is that coding agents provide a stepping stone toward artificial general intelligence, or AGI. This is the rationale for prioritizing coding, not a demonstrated result about reaching AGI.

### Use a benchmark as a development compass

[8:04](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=484s)

A busy repository creates an evaluation problem before it guarantees progress. Silen recalls thousands of early pull requests and a cadence of one pull request every two hours. The team lacked a clear way to tell whether a proposed change was good or how to test it. Evaluating changes consumed time and money without providing a reliable direction for development.

The response was to build a **benchmark** that could indicate whether the agent was improving. Category-level results make it possible to examine different kinds of capability rather than relying only on a general impression. Hackathon participants use the same evaluation tool. The team also reports running benchmark tests in its continuous integration pipeline across different open-source agents for the preceding couple of months.

The plotted benchmark success rate trends from about 35% to 55% over August, according to Silen, who calls the results noisy, messy, and unsuitable for a research paper. The chart preserves both observations: jagged individual series and a rising dashed trend. It is presented as a signal for directing development, not a general success rate that can be assumed for agents on other tasks.

![Line chart of success rate (%) over time, with several jagged agent series and a rising red dashed overall trend.](https://ai.engineer/api/talk-essay-media/452474a395a272fe0d5a9344dd953963cf9e30f3a47aa26378aa8e9946f99186)

Benchmark success rates fluctuate while the dashed overall trend rises.

### Two ways an agent can cause harm

[9:15](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=555s)

As agents gain capabilities, trust and reliability become more consequential. The first failure mode is **prompt injection**, which Silen identifies through OWASP’s attention to the problem. An agent visits a website and encounters malicious content. The language model treats that content as something it should obey, redirecting its behavior. The dangerous transition is from reading external material as data to accepting it as instructions.

The next failure does not require malicious instructions. Silen calls this category “innocently malicious”: an agent causes harm while attempting a legitimate task. In the displayed incident report, a user asked an open-source agent to delete JSON files within one specific directory. The reported result was deletion of JSON files across the laptop. The requested operation was deletion in both cases; the consequential error was expanding its scope beyond the requested folder.

![Post mentioning OpenInterpreter says a request to delete JSON files in one folder instead deleted all JSON files on a laptop; a terminal photograph appears below.](https://ai.engineer/api/talk-essay-media/dfcb7bbe438efaa6bfdbd2c5cfff52dc70da382ad8306f9759663d9399ab2986)

A displayed post reports JSON-file deletion extending beyond the requested folder.

That incident leads into the question of permissions. Silen argues that agents able to do what humans can do will need **root access**. This is his position on the access required for broad autonomy. In that vision, increasingly powerful access makes mistakes more consequential, so capability development cannot be separated from safety.

The team reports working on safety research and a paper addressing some of these problems. No concrete solution is presented here. The commercial concern is whether users can trust an agent with actions whose consequences extend beyond the conversation.

Silen sets 100% rather than 99% reliability as the requirement for commercial use; he is stating a target, not reporting an achieved result. One wrongly sent email could cost a contract or a lead. That example makes the stakes specific: even a seemingly small action can have a business consequence that an overall success rate obscures. Safety is therefore part of the team’s stated goal of building digital AGI that augments humanity.

### Fund the team behind the generalist agent

[11:17](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=677s)

The closing announcement returns to the community and its shared interest in advancing agent capabilities. The team announces a $12 million investment from Redpoint Ventures. It presents the investment as support for turning the agent vision into a working reality.

The team says Redpoint’s commitment to remaining open source was a reason for choosing the investor. The intended next step is to grow the team, with an invitation to contact its members and join the Discord community. The closing objective remains ambitious and explicit: bring more people into building what the team wants to become the world’s best open-source generalist agent.

## Resources

- [AutoGPT source repository](https://github.com/Significant-Gravitas/AutoGPT) — Source code and setup guidance for AutoGPT, with the original Classic agent separated from the later platform.
- [Agent Protocol](https://github.com/agi-inc/agent-protocol) — A common agent interface for creating tasks, executing steps, exchanging artifacts, and connecting compatible evaluation and development tools.
- [Testing Language Model Agents Safely in the Wild](https://arxiv.org/abs/2311.10538) — Silen Naihin and colleagues propose monitoring agent actions to constrain unsafe behavior during tests on the open internet.
- [The State of AI Engineering](https://www.ignorance.ai/p/the-state-of-ai-engineering) — An account of the inaugural AI Engineer Summit, including AutoGPT’s announced $12 million investment from Redpoint Ventures.

## Slides

### Octopus on mountains

A smiling octopus sits on snowy peaks as Silen argues that people can work faster and better with less time and stress.

[0:35](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=35s) [A smiling octopus sits atop illustrated snowy mountains.](https://assets.ai.engineer/sites/aiecode2025/sha256/ad75cd5e33a4e0516a74bbc2498d53f2bb1e5cd91e952016bbeefecca35409ff.webp)

AUTOGPT

### Lead generation spreadsheet

The speaker uses a lead generation spreadsheet to illustrate hours of searching and copying information between websites, then asks whether chatting could produce the same filled-out spreadsheet.

[0:58](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=58s) [An Excel spreadsheet lists companies, dates, locations, founders and websites.](https://assets.ai.engineer/sites/aiecode2025/sha256/5c15406c66c83cf79dba301ff0a432d8bb87c0243864ed25cb64eb708025904b.webp)

Name of the company | Creation date | Twitter Joined da | Country | CEO/Founder | Web
Naritiv | Jun '12 | October '2010 | Los Angeles | Dan Altmann | http://naritiv.com/
FameBit | Oct '13 | April '2013 | Santa Monica | David Kierzkowski | http://famebit.com/
TOTEMS | Feb '12 | September '2011 | London | Gabriel Hubert | http://totems.co/
Popular Pays | Apr '13 | March '2013 | Chicago | Corbett Drummey | http://www.popularpays.com/
NMRKT | Dec '11 | November '2010 | New York City | Julia Jacobson | http://nmrkt.com/
Snapwire | Aug '12 | August '2012 | Santa Barbara | Chad Newell | http://snapwi.re/
Revfluence | Oct '14 | | San Francisco | Anand Kishore | http://www.revfluence.com/
NeoReach | Apr '14 | | Stanford | Jesse Leimgruber | http://neoreach.com/
Captiv8 | Mar '15 | | San Francisco | Taz Patel | http://www.captiv8.io/
Queue | Nov '13 | October '2013 | San Francisco | Joe Sanchis | http://www.queueat.com/
Buzzoole | Nov '12 | October '2012 | Naples | Gennaro Varriale | http://www.buzzoole.com/
AlphaRank | May '15 | | San Francisco | Brian Ley | http://www.alpharank.io/
VidFall | May '14 | | Manchester, NH | Joel Robinson | https://vidfall.com/
Contigo.io | Mar '11 | May '2012 | San Francisco | Brian Ley | http://www.contigo.io/
GrapeStory | Apr '15 | | New York City | Gary Vaynerchuk | http://www.grapestory.co/
Blast Innovations | Feb '15 | | San Mateo | Kerstin Karu | www.blastbuzz.com/
MuseFind | Nov '14 | | New York City | Jennifer Li | http://musefind.com/
Foray Collective | Nov '13 | | Los Angeles | Tiana Haraguchi | http://www.foraycollective.com/
Activatr | Apr '15 | | New York City | Leon Lyazidi | http://activatr.com/
FindTheRipple | Mar '15 | May '2013 | London | Gianluca Angelici | http://www.findtheripple.com/
Yappee | Aug '14 | November '2013 | Solana Beach | Kevin Lee | http://www.yappee.com/
Style Coalition | Aug '12 | | New York City | Yuli Ziv | http://stylecoalition.com/
PopShorts | Mar '15 | | Los Angeles | Rob McCarty | http://www.popshorts.com/
Waypoint Media | Mar '15 | September '2013 | New York City | Christian Bingham | http://www.waypointmedia.com/
Sociolus | Oct '12 | November '2011 | London | Dinos Hamalis | http://www.sociolus.com/
Ader | Dec '14 | | Boston, MA | Justin Warden | http://getader.com/

Interface text:
[illegible] - Microsoft Excel
File | Home | Insert | Page Layout | Formulas | Data | Review | View | ABBYY FineReader 12
Paste | Calibri | 11 | General | Conditional Formatting | Format as Table | Cell Styles | Insert | Delete | Format | Sort & Filter | Find & Select
A1 | Name of the company
Sheet1 | Sheet2 | Sheet3
Ready | 100%
AUTO GPT

### Repetitive lead research

To fill a lead spreadsheet, the speaker describes repeatedly searching Google and LinkedIn and copying results back into the sheet for hours.

[1:13](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=73s) [A collage repeats LinkedIn logos, web searches and copy-and-paste keys.](https://assets.ai.engineer/sites/aiecode2025/sha256/05c90a81907dbe68dea6b179c7ebb9bb2039bd99a2637532ead982317dd7d967.webp)

Copy
Paste
Copy
Paste
Copy
Paste
Copy
Paste
Paste
Paste

in
in
in

AutoGPT

### Chat as an alternative to repetitive work

The speaker imagines using chat to produce a filled-out lead spreadsheet instead of repeatedly searching and copying information.

[1:33](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=93s) [A message input field on an AutoGPT presentation background.](https://assets.ai.engineer/sites/aiecode2025/sha256/a0e656a2d25de3b6829394f6355f6e95488a894fff85633ab184cf85e3c919aa.webp)

AUTOgpt

Interface text:
Send a message

### The unread email problem

An inbox badge shows 14,359 unread messages. The speaker asks whether chatting could replace hours spent sending variations of the same email.

[1:37](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=97s) [An email icon with a large unread-message badge.](https://assets.ai.engineer/sites/aiecode2025/sha256/6a91aacf15b6df406decc90bbac6c0d1637e5c7cf4ab9254fb78da98eed63a6f.webp)

14,359

AutoGPT

### Software development

The speaker contrasts lengthy app development and copying from ChatGPT with the possibility of building apps by chatting.

[2:21](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=141s) [A person typing on a laptop displaying a code editor.](https://assets.ai.engineer/sites/aiecode2025/sha256/f1aa2f822e23b0c1be347653bc4bf56d8e132c4b8483e0d9df100f2be63a23a2.webp)

[illegible]

AutoGPT

### AutoGPT in the press

Four outlets framed AutoGPT as an AI advance and a potential next big thing. The speaker links the coverage to its rapid rise to 100,000 repository stars.

[3:24](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=204s) [Four headlines from Fox News, TechCrunch, Mint, and Yahoo Finance about AutoGPT.](https://assets.ai.engineer/sites/aiecode2025/sha256/0f4579cf854568472a9b836644a9fa87b0564776f50c49424ffdae77faad587a.webp)

Fox News
What is Auto-GPT? Another step in the advancement of AI

TechCrunch
What is Auto-GPT and why does it matter?

Mint
Meet AutoGPT, the autonomous GPT-4 tool revolutionizing AI

Yahoo Finance
ChatGPT On Steroids? Why Auto-GPT Is The Next Big Thing In AI

AUTOgpt

### AI agents gain attention

The headlines place Bill Gates’s prediction alongside agent research: Voyager plays Minecraft, AutoGen coordinates multiple agents, and self-reflecting agents perform better in changing environments.

[3:28](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=208s) [Headlines from Reuters, NVIDIA Blog, VentureBeat, and Stanford HAI about AI agents.](https://assets.ai.engineer/sites/aiecode2025/sha256/43b40e8fa9cf4ac3dc17d09d56b52f7baaecbdf1e5aed715afa87c62323766ae.webp)

Reuters
Bill Gates says top AI agent will replace search, shopping sites

NVIDIA Blog
A Mine-Blowing Breakthrough: Open-Ended AI Agent Voyager Autonomously Plays ‘Minecraft’

VentureBeat
Microsoft's AutoGen framework allows multiple AI agents to talk to each other and complete your tasks

Stanford HAI
AI Agents that “Self-Reflect” Perform Better in Changing Environments

AutoGPT

### AutoGPT open-source repository

AutoGPT’s repository shows 150,000 stars and 462 contributors. The speaker thanks the people behind those numbers for their interest, pull requests and issues.

[4:28](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=268s) [AutoGPT repository page showing 150,000 stars and 462 contributors.](https://assets.ai.engineer/sites/aiecode2025/sha256/77d92ad24902fe3308085883fb1a12d809ebf605be2f1c48a2450acaf2fb44a6.webp)

AutoGPT

An experimental open-source attempt to make GPT-4 fully autonomous.

agpt.co

Interface text:
Public
Edit Pins
Watch 1.5k
Fork 33.7k
Starred 150k
master
110 branches
19 tags
MissingDLL and SilenNaihin: jarvis3 entering the arena (#5630)
2d865cc now
4,526 commits
.github — Frontend ci (#5579) — 2 days ago
arena — jarvis3 entering the arena (#5630) — now
autogpts — AutoGPT/plugins: Support full parameter defs for plugin commands — yesterday
benchmark — Fix password generator (#5581) — 2 days ago
docs — forge - tutorial 3 adding images — 2 weeks ago
frontend — Introduce TestOption Enum for Enhanced Test Selection Clarity (#5586) — 2 days ago
.flake8 — Add .flake8 — 3 weeks ago
.gitattributes — updated git attributes — yesterday
.gitignore — Granger entering the arena (#5401) — last week
.gitmodules — Fix Auto-GPT CI cassette submodule — last month
CLI-USAGE.md — Removed additional refs to frontend — 3 weeks ago
python
ai
artificial-intelligence
openai
autonomous-agents
gpt-4
MIT license
Activity
150k stars
1.5k watching
33.7k forks
agbenchmark-v0.0.10 — Latest — 3 weeks ago
Contributors 462

### Community matters

AutoGPT credits its community and commits to growing and empowering it. The slide shows 47,224 Discord members, 150k GitHub stars and 464 contributors.

[5:57](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=357s) [Community slide highlighting AutoGPT's members, stars, and contributors.](https://assets.ai.engineer/sites/aiecode2025/sha256/368e6e364ce115014bc1670ab8351e9ecced3c67f5278dc29bd69bf1c3280b96.webp)

Community matters.

🌟 AutoGPT: the heart of the open-source agent ecosystem

AutoGPT 47224 members
Stars 150k
Follow @auto_gpt
License MIT

Contributors 464

AutoGPT

Interface text:
README.md

### Forge for agent creators

AutoGPT built Forge as a standardized template to help creators develop agents. An anvil sends arrows toward the AutoGPT logo and two illustrated people.

[6:31](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=391s) [An anvil and hammer with arrows pointing toward the AutoGPT logo and two illustrated people.](https://assets.ai.engineer/sites/aiecode2025/sha256/122183532607039b04a51db1d2748da307276ddc7cb7ee4e33d4247f5ab31025.webp)

AutoGPT

## Full transcript

[0:00](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=0s) [upbeat music] Thank you, San Francisco, for the warm welcome.

[0:18](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=18s) I'm Toran, the creator of AutoGPT, and I'm excited to show you all what the brilliant minds at AutoGPT have been working on over the past months. I'm going to hand off the stage now to Silen, one of our founding AI engineers.

[0:35](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=35s) Thank you, Toran. Thank you. There you go. Thank you, Toran. Um, I want to talk about something that I think not many of you realize. I didn't realize this for a long time.

[0:48](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=48s) We're not achieving the peak of our potentials.

[0:51](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=51s) We can all work faster, we can work better, and we can do more with less time and less t- and less stress.

[0:59](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=59s) Let's take this spreadsheet, for example. I don't know about you, but I've stared at this interface for hours on end, and, um, I'm sick of it. How would you go about filling out this spreadsheet, all right?

[1:09](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=69s) It's lead generation. Name of the company, you got the links. Uh, what you'd probably do

[1:16](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=76s) is go on Google, you search, copy/paste, maybe go on LinkedIn, copy/paste, back to Google, over and over and over again for hours, going back to the same interface, going back to the same websites.

[1:28](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=88s) But what if instead of all that, you can just chat and you get the same end result, a filled out spreadsheet with all the leads? All right, let me give you another example.

[1:39](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=99s) We all have unread messages, right? Not because we're lazy, allegedly, but because we're overwhelmed.

[1:47](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=107s) Now, how would you go about cleaning out your inbox?

[1:50](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=110s) You'd sit there for hours and hours sending the same variation of the same email.

[1:56](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=116s) But what if you could just chat? Last example, I promise. Actually, uh, these emails will now be leads in your inbox, uh, instead of, uh, just unread emails. Last example.

[2:12](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=132s) Say you're a company or a developer. You spend millions of dollars developing apps that take weeks, months, sometimes even years,

[2:22](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=142s) sitting there copy/pasting anyways because you're probably using ChatGPT. I know I am copy/pasting. But what if instead of all that effort, you just chat?

[2:33](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=153s) I think you get the point. There's a reason you've heard of AutoGPT.

[2:39](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=159s) AutoGPT inspired the minds of millions. It gave hope to what a world could look like where we all reach our full potential. The light at the end of the tunnel.

[2:49](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=169s) You could see the sparks of digital artificial intelligence.

[2:53](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=173s) And in this world, everyone goes from using their minds mostly to execute menial tasks with only 10% of their brains being used for creative work

[3:05](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=185s) to becoming creative masterminds, orchestrating the peak potential of their lives. And in this world, we're all AI engineers, whether you know it or not.

[3:16](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=196s) And people have noticed AutoGPT was the fastest repository to 100,000 stars. Every major new- news network picked up on this.

[3:26](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=206s) Everyone understands what the potential of this is, and it kicked off a whole new field of development, a whole new paradigm of augmenting humans to give them time back and live a more stress-free life.

[3:40](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=220s) And even the major players in the space all realized how big of a deal it is and work-- and now work on these agents.

[3:48](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=228s) And so I want to hand off to the primary open source developer at AutoGPT to talk a, uh-- to talk a little bit more about the open source repo.

[4:01](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=241s) Thank you, Silen. Thank you, Silen, and hello, San Francisco. [audience applauding] [audience cheering]

[4:13](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=253s) I think all of us being here is a real testament to the power of open source. And on that note, we have some really exciting news to share.

[4:24](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=264s) Because just last week, our open source repo, AutoGPT, hit 150,000 stars on GitHub. [audience applauding] [audience cheering]

[4:45](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=285s) Of course, metrics are fun, but to me it is so much more than just a number.

[4:52](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=292s) It is the 150,000 people who took an interest in what we're doing and decided to click that button. So if you star our repo, then thank you.

[5:04](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=304s) It is also the 460 plus contributors who took the-- their time and effort submitting thousands of pull requests and issues

[5:15](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=315s) in the process, and to all of them as well, thank you so much.

[5:21](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=321s) It is also the 47,000 members of our online community

[5:28](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=328s) and all the interesting and insightful interactions that they've given us.

[5:33](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=333s) It's been a wild ride at times, but it, it has allowed us to do and learn so much in the past six months.

[5:43](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=343s) And I'm extremely excited for what is to come based on that.

[5:52](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=352s) Now, I've already said it, but we could not have done this without our community, and community matters.

[5:59](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=359s) So we are committed to fostering, to growing, and to empower this community

[6:07](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=367s) And to build the future together. And I'll hand it back to Silen to tell you what that means

[6:19](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=379s) Thank you, Poots. And we haven't stayed stagnant since the open source agent originally came out. We've continued to work on it, and we've continued to improve its capabilities and implement the latest cuttin- cutting-edge research.

[6:30](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=390s) But we've also been working on some other things. To show our commitment to the agent space and the open source ecosystem, we built a forge, which is a template for any agent, for any agent creator to have a better time to develop their agents with a standardized template.

[6:46](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=406s) We also built a dev tool UI to easily interact with and iterative- iteratively improve your agent using an intuitive interface.

[6:54](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=414s) All of these tools are built on top of the Agent Protocol from the AI Engineer Foundation and other industry standards to maximize compatibility and interoperability. Anyone who implements this protocol can use our benchmark, frontend dev tool, and other offerings built on top of this protocol.

[7:11](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=431s) And while this dev tool template is in beta or in alpha, uh, it has served, uh, our participants of the current hackathon we're running, where we have $30,000 in cash on the line.

[7:21](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=441s) And we've learned a lot from this. We've received a lot of great feedback. We've received a lot of bug fixes, uh, and insights that we're going to take into the future.

[7:29](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=449s) One of those insights is that code is king.

[7:35](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=455s) We've realized that coding agents are the fundamental agents, uh, of the world.

[7:42](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=462s) Let me tell you, the digital fabric, the fundamental digital fabric is code.

[7:49](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=469s) Our goal is to build a general a-- generalist agent, yes, but code is the s- is the stepping stone to AGI.

[7:56](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=476s) A motivated coder can get anything done, uh, except for getting a bed frame. [laughs]

[8:04](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=484s) Another thing that we've learned over time is that without a compass, you don't know where you're going. You know, at the start of the repo where we're getting thousands of pull requests, and

[8:14](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=494s) that's a pull request every two hours. We had no way to know whether the pull requests were good and how do we even test these pull requests. We didn't have a real direction.

[8:24](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=504s) It took time to test these, and it was unnecessarily costly. And so we created a compass. We created a benchmark to direct the development of the open source repo and quantitatively know if we were improving.

[8:36](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=516s) It's an easy way to know if your agents are improving down different categories, and people are currently benefiting from this for the virtual hackathon.

[8:44](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=524s) And this is just cool. We've been running this in our CI pipeline for the past couple months on different open source, uh, agents within the ecosystem. And what the tests have shown is that we're on the brink of something special.

[8:58](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=538s) These agents have showed continual improvement, and don't worry, I wouldn't put this in a research paper. The-- It's very noisy. It's very messy. But there is a continuous trend from thirty-five to fifty-five percent.

[9:08](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=548s) Uh, this is just a graph of the success rate on the benchmark over time, over the month of August.

[9:15](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=555s) Another thing that we're committed to is safety.

[9:20](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=560s) As the ecosystem grows and as the capabilities of agents increase,

[9:25](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=565s) there's always questions of trust and reliability, and these are problems that AutoGPT is committed to. One of these problems is prompt injection, which will always be there. OWASP, uh, one of the big security organizations, ha- has talked about this and said this is one of the big problems that not just language models face, but also agents.

[9:44](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=584s) It's essentially when agents visit a website, which all agents need to do, and, uh, the website has something malicious. Um, and then the LLM is like, "All right, I need to be doing that now."

[9:55](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=595s) And you can see that there, uh, in this example.

[9:59](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=599s) Then there's this other, uh, category that I like to call, uh, innocently malicious, where, uh, agents are just bad sometimes. It's the truth. And in this example behind me, uh, this person asked, uh, an open source agent to delete all the JSON files within a directory, a specific one, and the agent ended up deleting all the JSON

[10:21](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=621s) files on a laptop. And this is gonna continue to be a problem. If we want agents to do the things that humans can do, they will need root access.

[10:30](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=630s) And so within AutoGPT, we're committed to and think about these problems extensively, and we've been working on an, a research paper to solve some of these issues. Um, and in order for agents to be commercially viable and trusted, these safety problems need to be solved.

[10:45](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=645s) You can't have a ninety-nine percent success rate. It has to be one hundred percent. That one email that's sent could be a lost contract or a lost lead. And so this is fundamental, not just to the development of the open source agent, but to all agents out there.

[11:00](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=660s) After all, our end goal is a digital AGI to augment all of humanity. And I'm gonna invite Craig to announce some exciting news regarding some developments with AutoGPT.

[11:17](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=677s) Hello. So it's been a wild journey from zero to here in six months,

[11:27](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=687s) and we keep stressing this because it's so important to us. We're only here because of our community, because of that shared passion in pushing the frontier of what AI agents can do.

[11:40](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=700s) So we're really excited to announce that Redpoint Ventures has invested twelve million dollars in turning this vision to a reality. [audience applauding]

[11:56](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=716s) Now, this isn't just funding. This is that, uh, them showing their deep belief in our mission and their dedication to open source. That's why we went with them, because they are so dedicated to staying open source, and that's really important to every single one of us working on this project.

[12:17](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=737s) Now, this is where we need you. With this funding, we wanna grow our team and add more passionate individuals.

[12:26](https://www.youtube.com/watch?v=ZYoZSU58m_Y&t=746s) So join us. Message us. Join our Discord community, and let's all help make the world's best open source generalist agent. Together, we can redefine the future of work. [audience applauding] [upbeat music]
