World's First AGI Agent SHOCKS the Entire Industry! (FULLY Autonomous AI Software Engineer Devin)

Summary

Cognition Labs introduces Devon, the world's first AI software engineer, capable of autonomously tackling complex engineering tasks. Devon demonstrates its abilities by benchmarking API performance, debugging, building websites, and even fine-tuning AI models. The AI's proficiency in using developer tools and learning from documentation showcases the potential for AI to revolutionize software engineering, offering a glimpse into a future where AI assistants like Devon could automate and enhance various aspects of the profession.

Takeaways

• 🚀 Introduction of Devon, the first AI software engineer, capable of performing complex tasks like a human engineer.
• 🛠️ Devon can create a step-by-step plan, build projects, and use tools such as a command line, code editor, and browser.
• 📚 Devon autonomously learns by reading API documentation and other technical materials to solve problems.
• 💡 Devon has the ability to debug code by adding print statements and fixing bugs based on error logs.
• 🌐 Devon can build and deploy fully styled websites, showcasing its capabilities in web development.
• 📈 Devon has successfully passed practical engineering interviews and completed real jobs, demonstrating its real-world applicability.
• 🤖 The development of Devon represents significant advancements in AI reasoning, long-term planning, and autonomous task execution.
• 🎥 A video from 6 months prior discussed the concept of autonomous AI agents running software businesses, which is now becoming a reality with Devon.
• 🔧 Devon is equipped with common developer tools within a sandboxed computer environment, allowing it to perform tasks securely.
• 🏆 Devon outperforms other AI models in benchmarks for resolving real-world GitHub issues, indicating its superior problem-solving skills.
• 🌟 The potential future scenario where autonomous AI agents like Devon could run businesses, performing tasks and customer service without human intervention.

Q & A

What is Devon and what makes it unique?

Devon is the world's first fully autonomous AI software engineer developed by Cognition Labs. It is unique because it can perform complex engineering tasks, learn over time, and fix mistakes. Devon is equipped with common developer tools and can operate within a sandboxed computer environment, making it capable of end-to-end development and deployment of applications.

How does Devon tackle a problem?

Devon approaches a problem by first creating a step-by-step plan to tackle the issue. It then builds a project using the same tools a human software engineer would use. If it encounters an error, Devon adds debugging statements, reruns the code, and uses the error logs to fix the bug.

What are some real-world applications of Devon?

Devon has been used to complete real jobs on Upwork, fine-tune a 7B llama model, set up a computer vision model, and fix bugs in existing software. It has also been used to implement a game of life, improve user experience in an open-source tool, and autonomously learn from a blog post to generate a desktop background image.

How does Devon's performance compare to other AI models in solving real-world GitHub issues?

In a benchmark for resolving real-world GitHub issues, Devon achieved a 13.86% success rate, which is significantly higher than other models like GPT-4, making it around 7 times more effective than GPT-4 in this context.

What kind of support does Devon provide to human engineers?

Devon can assist human engineers by taking on tasks such as running commands, tracking their status, fixing bugs, writing test cases, and improving user experience in tools. This allows engineers to focus on more interesting problems and achieve more ambitious goals.

How does Devon's learning process work?

Devon learns by reading documentation, running code, and understanding the context of tasks. It can recall relevant context at every step and adapt its approach based on the information it gathers, allowing it to learn from its experiences and improve over time.

What is the potential impact of Devon on the software engineering field?

The introduction of Devon could revolutionize the software engineering field by automating complex tasks, reducing the time taken to solve problems, and enabling engineers to work on more innovative projects. It could also lead to the creation of new job roles that focus on managing and optimizing AI software engineers like Devon.

How does Devon handle versioning issues?

When faced with versioning issues, Devon updates the code to make it compatible with the required versions. It also uses tools like pip to manage dependencies and ensure that the project runs smoothly.

What is the significance of Devon's ability to use a browser?

Devon's ability to use a browser is significant as it allows it to access API documentation, learn how to integrate with various APIs, and gather information from the internet to assist in problem-solving and project development.

How does Devon's deployment of a website showcase its capabilities?

Devon's deployment of a website with full styling demonstrates its ability to not only code but also to create visually appealing and functional end-products. It shows that Devon can understand design requirements, implement them in code, and deploy the final product, just like a human developer.

What is the future potential of autonomous AI agents like Devon?

The future potential of autonomous AI agents like Devon is vast. They could lead to the automation of various aspects of business operations, from customer service to product development. As they become more advanced, they could potentially run entire businesses, allowing humans to focus on higher-level tasks and innovation.

Outlines

00:00

🤖 Introduction of Devon, the AI Software Engineer

The video introduces Devon, the first AI software engineer developed by Cognition Labs. Scott, from Cognition AI, showcases Devon's capabilities by asking it to benchmark the performance of a llama and different API providers. Devon demonstrates its problem-solving skills by creating a step-by-step plan, using tools like a command line, code editor, and browser to tackle the task. It encounters an error, adds a debugging statement, and fixes the bug using logs. The video emphasizes the advancements in AI's reasoning and long-term planning, and invites viewers to try Devon for real-world tasks.

05:01

🛠️ Devon's Problem-Solving and Debugging Skills

The paragraph details Devon's ability to handle complex tasks such as setting up a computer vision model for a client on Upwork. Despite encountering versioning issues, Devon updates the code to resolve them and proceeds to load and import packages, downloading images to run the model. Devon also performs print line debugging to understand data flow and corrects the code accordingly. The paragraph highlights Devon's persistence and problem-solving approach, ultimately delivering a report with sample images and a detailed explanation of its work.

10:03

🎮 Devon Assists in Game Development and AI Training

Devon's versatility is showcased as it assists in implementing the game of life and fine-tuning a 7B llama model. For the game, Devon creates a React application, writes code, and deploys it, making adjustments based on user feedback. In the AI training task, Devon fine-tunes a large language model, overcoming CUDA issues, and successfully running the training job. The paragraph emphasizes Devon's ability to learn and adapt to new tasks, such as understanding and applying a fine-tuning method to a language model.

15:06

🔧 Devon Fixes Bugs and Improves User Experience

The paragraph describes Devon's role in fixing bugs and improving user experience for various projects. It helps enhance an open-source tool's UX by understanding the code and making necessary changes. Devon also assists in debugging a Python algebra system, identifying and fixing an issue with log calculations. Another engineer shares his experience with Devon, who helps in writing and expanding test cases for an open-source repository, ultimately finding and fixing a bug. The paragraph highlights Devon's ability to understand and manipulate code, debug issues, and enhance software quality.

20:08

🚀 Devon's Autonomous Learning and Business Potential

The video script ends with a scenario where Devon autonomously learns from a blog post to generate a desktop background image. It also discusses the potential of Devon as an AI software engineer in a business context, where it could handle all aspects of a custom automation business, from customer interaction and solution brainstorming to software development, part ordering, and customer service. The scenario illustrates the potential for AI to revolutionize business operations, requiring minimal human intervention and leading to significant efficiency and profit.

Keywords

💡Devon

Devon is the first AI software engineer introduced by Cognition Labs. It is designed to autonomously perform complex software engineering tasks, such as benchmarking, debugging, and deploying websites. In the video, Devon is showcased as a capable assistant that can tackle real-world programming challenges, learn from documentation, and even complete jobs on platforms like Upwork. The AI's ability to use tools such as a shell, code editor, and web browser demonstrates its advanced capabilities in software development.

💡AI Software Engineer

An AI software engineer refers to an artificial intelligence system, like Devon, that is capable of performing the tasks typically associated with a human software engineer. This includes coding, debugging, project management, and even learning new technologies or programming languages. The concept is significant because it represents a leap in AI capabilities, suggesting that AI can now operate with a level of autonomy and complexity previously reserved for human professionals.

💡Benchmark

Benchmarking is the process of evaluating the performance of a system or component by comparing it to a standard or other similar systems. In the context of the video, Devon is shown benchmarking the performance of a llama and different API providers, which involves testing and analyzing their efficiency and effectiveness. This is crucial for ensuring that the software or systems being developed meet the required performance standards and for identifying areas of improvement.

💡API Providers

API, or Application Programming Interface, providers are entities that supply the protocols and tools for building software applications. They allow different software to communicate with each other. In the video, Devon interacts with various API providers to understand their documentation and integrate with their services, which is a common task for software engineers when developing applications that rely on external data or services.

💡Debugging

Debugging is the process of identifying and fixing errors or bugs in a computer program. It is a critical part of software development and maintenance. In the context of the video, Devon demonstrates its debugging skills by adding print statements to track data flow and identify issues in the code. This ability is essential for AI software engineers to ensure that the code they produce is error-free and functions as intended.

💡Upwork

Upwork is a global freelancing platform where businesses and individuals can find independent contractors for various projects, including software development. In the video, Devon's capabilities as an AI software engineer are highlighted by its successful completion of real jobs on Upwork, demonstrating its practical application in the freelance market.

💡Autonomous AI Agents

Autonomous AI agents are artificial intelligence systems that can operate independently, making decisions and executing tasks without continuous human intervention. In the video, the concept of autonomous AI agents is discussed in the context of Devon's capabilities, suggesting a future where AI can run businesses, satisfy customer orders, and create scripts autonomously.

💡Long-term Planning

Long-term planning refers to the ability to strategize and make decisions that take into account future outcomes and goals. For AI software engineers like Devon, this involves the capacity to plan and execute complex engineering tasks that require a series of decisions and actions over an extended period. This is a critical skill for managing projects and ensuring that they are completed successfully and efficiently.

💡Code Editor

A code editor is a software application used for writing and modifying computer code. It typically offers features like syntax highlighting, auto-indentation, and code suggestions to aid programmers. In the context of the video, Devon utilizes its own code editor to write and debug code, which is a standard tool for software engineers and a testament to Devon's capabilities as an AI software engineer.

💡Sandboxed Environment

A sandboxed environment is a secure and isolated space within a computer system where programs can run without affecting the host system. It is used to test and develop software safely, as any changes or potential security issues are contained within the sandbox. In the video, Devon operates within a sandboxed computer environment, equipped with common developer tools, which allows it to perform software engineering tasks without risking the stability or security of the host system.

💡GitHub Issues

GitHub Issues is a feature of the GitHub platform that allows users to track and manage tasks, enhancements, and bugs for their projects. It is a collaborative tool that facilitates communication between developers and helps in organizing and prioritizing work. In the context of the video, Devon's ability to resolve real-world GitHub issues demonstrates its practical application in software development and its capacity to interact with widely used development tools and platforms.

Highlights

Cognition Labs introduces Devon, the world's first AI software engineer, capable of performing complex engineering tasks.

Devon can benchmark performance, create step-by-step plans, and build projects using tools like a human software engineer.

Devon has its own command line, code editor, and browser to execute tasks and learn from API documentation.

The AI software engineer can debug code by adding print statements and fixing errors based on logs.

Devon can build and deploy fully styled websites, showcasing its capabilities in web development.

Advancements in reasoning and long-term planning have made it possible for AI like Devon to tackle complex problems.

Devon has successfully passed practical engineering interviews and completed real jobs on platforms like Upwork.

The AI agent autonomously solves engineering tasks, including setting up computer vision models and fine-tuning large language models.

Devon demonstrates the ability to learn from blog posts and apply the knowledge to generate desktop background images.

Cognition AI's development of Devon represents a significant leap in AI capabilities, surpassing current tools available to the general public.

Devon's performance on benchmarks shows it is seven times more effective than GPT-4 at resolving real-world GitHub issues.

The AI software engineer can autonomously learn and improve code, fixing bugs and edge cases not covered in the original documentation.

Engineering teams can achieve more ambitious goals with Devon's assistance, as it can handle complex tasks requiring thousands of decisions.

Devon is equipped with common developer tools within a sandboxed computer environment, simulating a human developer's workflow.

Cognition Labs' creation of Devon signifies rapid progress in AI, with capabilities beyond what was anticipated just six months prior.

The potential applications of Devon include running businesses autonomously, providing custom automation solutions, and handling customer service.

Devon's ability to learn and apply new skills suggests a future where AI agents could become commonplace in various industries.

Cognition Labs has raised $21 million in series A funding led by Founders Fund to further develop Devon's capabilities.

Transcripts

so cognition Labs just drops this Devon

the first AI software

engineer hey I'm Scott from cognition Ai

and today I'm really excited to

introduce you to Devon the first AI

software engineer let me show you an

example of Devon in

action I'm going to ask Devon to

Benchmark the performance of llama and a

couple different API

providers from now on Devon is in the

driver's

seat first Deon makes a step-by-step

plan of how to tackle the

problem after that it builds a whole

project using all the same tools that a

human software engineer would use Devon

has its own command

line its own code

editor and even its own

browser in this case Devon decides to

use the browser to pull up API

documentation so that it can read up and

learn how to plug into each of these

apis here Deon runs into an unexp

[Music]

error Deon actually decides to add a

debugging print

statement reruns the code with the

debugging print statement and then uses

the error in the logs to figure out how

to fix the

bug finally Devon decides to build and

deploy a website with full styling as

the

visualization you can see the website

here

all of this is possible today because of

the advancements that we've made in both

reasoning and long-term planning it's a

really hard problem and we've only just

started but we're super excited about

the progress that we've made so

far in the meantime if you'd like to try

out Devon on your own real world tasks

send us a request below and we'd be

happy to forward it to Devon about 6

months ago I made this video about

autonomous AI agents now this was before

that terminology was quite as used you

we're hearing it more now back then it

was isn't quite as common so this was 6

months ago August 27th 2023 I'll play a

clip from this towards the end of the

video but the question I was asking that

video was how far away are we from a

situation where basically an autonomous

AI agent would be able to kind of run a

software business on your behalf satisfy

customers orders create little scripts

for them send those out Etc and you know

that video is very well received a few

people thought that's never going to

happen that's a million years away and

some people gave their kind of timelines

for that but I think overall a lot of

people said how Blown Away they were by

this concept of having an autonomous

worker that's able to generate money for

you kind of a big concept to think about

on many different levels not just for

yourself but how would having access to

something like that change the world

it's been about 6 months that was 6

months ago keep that keep that in mind

people were super interested some small

percentage was calling BS that was 6

months ago and this this is today today

we're announced to excite I can't even

read right now okay Wes calm down and

try that again today we're excited to

announce Devon the first AI software

engineer so Devon has successfully

passed practical engineering interviews

from leading AI companies and has even

completed real jobs on upwork so it does

the job interview it completes the jobs

demon is an autonomous agent that solves

engineering tasks through the use of its

own Shell Code editor and web browser so

let me do this so this thing isn't out

yet for everyone I am I am spamming the

crap out of anyone that can uh get me

access so I'm really hoping to have

access to this thing ASAP big props to

David Andre for breaking the story

that's where I saw it first the title

said world's first AGI agent yes this is

real I was like no it's not and I

clicked it anyways and yeah it kind of

seems like it just might be let me know

in the comments what you want to see

this thing do and watch these videos

that showcase its

abilities hey I'm Walden one of the

developers here at cognition AI we were

playing around with whether or not Devon

could start a side hustle on upwork so

here's actual real job from upwork where

the client wants to set up this computer

vision model which actually looks quite

interesting seems very difficult to set

up um I'm not sure how I would start

doing this but you know you give the

task to Devon and ask Devon to figure it

out and things just kick off Devon

immediately

goes ahead and you can see it sort of

starts setting up the repo it actually

runs into some issues here with the

versioning so if you watch how Devon

deals with

it deon's actually updating the code to

make these things

work he continues with this loading and

importing packages you can see that

actually downloads images from the

internet to run through the

model

but you can see here that there are

actually some issues that come across

however Devon knows how to handle these

things Devon kind of pushes through and

if you look closely Devon's actually

doing print line debugging

here where Devon is adding these

statements to track where the data

flows and Devon continues to do this

until Devon understands how everything's

working and actually then updates the

code with the fixes

after removing print line

statements Deon continues this pattern

of fixing code and running it again

until it runs the image model across all

these roads across the world and we can

ask for a report from

Devon at which point Devon sends over

some sample images of roads with damage

marked

out and a nice txt file explaining

Devon's work and the different kinds of

outputs of the model good job

de hi I'm Adon and today I felt like

playing the game of life so I asked Deon

to implement it for

me Deon started by creating a new react

application using the Shell and then it

started writing some code through its

editor after that it deployed the app

through netlify let's check it

up that seems nice um but there's a lot

more features which I want to add

so let's ask Deon to do this one at a

time I want the words Dev to be written

at the initialization screen instead of

it being

random then I want the word to be

slightly bigger and the frame rate to be

faster I also want him to fix a bug

where the screen gets freezed after 3

seconds let's see the progress dev has

made so far

you can see the diff and um the last

diff shows that Devon just fixed the bug

uh where the screen gets frozen after 3

seconds the seems reasonable to me so

let's move

on next I want Deon to increase the

frame rate after 10

seconds and also to make the website

responsive to different window

sizes also wanted to make it interactive

so that when when I click my mouse

somewhere it should spawn a new

block let's check out what Deon has made

so

far started with de which is what we

asked for and when I click something it

creates a new block as

well that's

fun um let's play around with

it well that goes my evening

hey guys today I'm going to show you an

AI training in

AI so here we're going to take the Cur

repo which is a fine-tuning method for

quantizing large language models we're

going to feed this repo to our agent

Devin and all we have to ask Devon is to

fine-tune a 7B llama model Devon clones

the repo

figures out how to run it using the

readme sets up all of the requirements

using

pip looks through all the

scripts and is able to start running the

training job there are a few hiccups

where Devon runs into some Cuda issues

which is to be expected with open source

repos but it's not a problem Devon looks

at the Nvidia environment

and figures out how to reinstall the

packages to make it

work after a few more runs figure out

the correct model

names Devon successfully gets the

training run

working here we see training proceeding

smoothly loss is going

down and

after few steps looks pretty good I tell

Devon to wait as the training job

runs after about an hour I come back ask

Devon hey how's the training going Devon

helps me look a few hundred steps are

done now and everything is still

proceeding

smoothly looks great thanks Devon for

helping me set up my training

run hey I'm Tony an engineer cognition I

helped build Devon and now Deon helps me

too today at work I wanted to run a

bunch of commands at once and be able to

track their status on one screen I found

an open source tool named impro to do

this here it is right here looks like it

all finished but the status is way too

vague I don't know which ones failed

they all just say

down I really want to improve the ux

here but I'm not familiar with the code

at all so I had Devin my AI software

engineer help me looks like this person

right here had the same issue so all I

gave Devon was the link to the issue and

asked Deon to fix it you can see me make

the request right here on the left let's

see what Devon did on the right we can

track deon's work and watch Devon jump

from tool to Tool first Deon Clon the

repository using the Shell then reads

the read me and an Editor to learn how

to sub the

code then goes back to the Shell to

install the required

dependencies Devon also opens up a web

browser

to take a look at the

issue now Devon starts

coding at some point Devon even opens up

some Rust documentation to debug a

compiler

error finally Deon finishes the task and

reports a summary of the changes that

were

made let's see the changes work I have

deon's code right

here looks like it worked the third

command succeeded I can even see the

status

codes here's all the code that Devon

wrote for this

change thanks

Devon hey I'm Neil and I wanted to show

you an example of Devon our AI software

engineer helping me fix a bug so I've

been using this repo called Senpai

Senpai is an algebra system written in

Python and I noticed this issue where

when you take the log of a fraction you

get Zoo which is a type of

infinity so that's definitely wrong but

instead of trying to figure this out

myself I just asked Deon to take a

look Devon immediately jumps in sets up

the repo and is able to reproduce that

same Zoo

output Devon then figures out the right

part of the code and adds print

statements um in order to figure out

what the cause of this issue

is

and we can see here that the cause is

that integer division leads to a zero

and then we take the log of zero So

based on that Devon's able to fix the

issue in The Code by replacing that

integer division with true Division and

then cleans up the debug

output and verifies that the result is

what we want and then Devon even runs

the test and the repo as well to make

sure nothing else is broken so that was

great um saved me a ton of time so thank

you

Dev hey I'm Andrew an engineer at

cognition and I wanted to share a pretty

amazing experience I had with Devon so I

maintained this big open source

repository uh which contains a lot of

different algorithms uh used for

competitive programming a lot of people

use it and a few weeks ago uh my friend

texted me that you know there was

actually a bug in one of the in one of

the implementations uh the

implementation wasn't quite right when

the inputs weren't uh weren't relatively

prime I kind of glossed over that case

when I was implementing it so I never

really thought about it so I implemented

a quick fix and then I thought that I

should test it but I actually never

really got around to writing any test

cases so I thought if I don't want to do

it I should just ask Devon to do it

instead so I gave Deon the repository

asked uh asked Deon to just check it out

and start working on it uh so Deon you

know found the right repository checked

it out you know found all files that are

in the repository and then I told Devon

what test case I wanted him to

write uh I just told Deon you know these

are the inputs and then try checking for

these conditions for me so Devon wrote

the test without too much trouble uh it

was Devon just looked around to

understand what exactly uh what exactly

the test should look like and what

exactly the interfaces were and with

this Devon ran the tests ran into a

quick hiccup which was a compile ER but

Devon is able to solve those very

effectively and just added an extra

include to fix that and then uh was done

writing this initial test so then I

asked de to actually expand the test a

little bit instead of just testing this

one input I wanted Deon to write test it

on all inputs so kind of the Brute Force

testing strategy I use this a lot in my

test and I just wanted Devon to

implement it so that I didn't have to

worry about it so de went and rewrote

the test function to use four n for

Loops but this time after Deon ran the

tests Devon actually found a test

failure now you know if the code were

correct there could be compilers in the

test but you know the test seemed really

pretty reasonable so there probably

shouldn't be a failure so Devon went and

tried to debug the program for me so

Devon here actually wrote uh actually

added a print statement to debug the

outputs uh and the inputs to the failing

test reran the tests and actually found

which case was wrong in this case these

are the inputs and then the return value

was actually 9 uh and the the code I'm

running actually should never really

return negative values so Deon realized

this and actually went looking in the uh

went looking in the code that we're

trying to test and actually added this

line of code that if extra less than

zero extra plus equals you know plus

equals something and in order to make

sure that the return value was actually

non-

negative so after fixing this Devon

actually reran the tests and now uh now

I can be confident that my code is

correct and I have some tests to prove

it thanks

Deon hey everyone my name is Sarah and

I'm going to show you how Devon our AI

software engineer can autonomously learn

from a blog post within a few minutes

Devon successfully generated this

desktop background image for me with my

name on it so all I had to do was send

this blog post in a message to Devon

from there Devon actually does all the

work for me starting with reading this

blog post and figuring out how to run

the

code in a couple minutes Devon's

actually made a lot of progress and if

we jump to the middle here you can see

that Devon's been able to find and fix

some edge cases and bugs that the blog

post did not cover for me and if we jump

to the end we can see that Devon uh

sends me the final result which I love I

also got two bonus images uh here and

here so uh let me know if you guys see

anything hidden in these so this is

cognition Labs website so that you've

raised a 21 million series a led by

Founders fund meet Devon the world's

first fully autonomous AI software

engineer Deon is a tireless skilled

teammate equally ready to build

alongside you or independently complete

tasks for you review with Devon

Engineers can focus on more interesting

problems and Engineering teams can

strive for more ambitious goals with our

advances and long-term reasoning and

planning Devon can plan and execute

complex engineering tasks requiring

thousands of decisions Devon can recall

relevant context at every step learn

over time and fix mistakes we've also

equipped down of common developer tools

including the Shell Code editor and

browser within a sandboxed computer

environment everything a human would

need to do their work what I find really

interesting about this stuff is

generally when you're building tools

aimed at developers um I mean generally

speaking you have less restrictions

versus something that's aimed at sort of

everybody so from what we've seen so far

because this seems much more powerful

than anything else that's available

right now to the general public being

able to learn unfamiliar technology so

like giving it a blog post and then it

knows how to do stuff I mean that was

the promise that's kind of what we

expected these things to be able to do

we've seen some of that like that in

context learning by large language

models we've seen it but this seems Next

Level certainly build and deploy apps

end to endend now on this channel we've

shown for example chat Dev and autogen

and stuff like that so this isn't new

but just everything about this seems

smoother faster more intuitive and more

what is even the word for this more

agentic like it just seems smarter and

more competent you tell it what to do

and it goes and does that so thewe bench

is a benchmark for you know can language

models resolve real world GitHub issues

and so here's kind of the results that

they have right here with Claude 2 GPT 4

Etc assisted and unassisted so looks

like yeah Cloud 2 4.8 unassisted GPT 4

1.74% unassisted and so Devon just

completely knocks it out of park so

that's

13.86% 13 Point let's say just call it

14% so what is that 7x better than GPT 4

looks like Cloud 3 hasn't been tested

yet is isn't on The Benchmark quite yet

so I am tirelessly trying to wrestle up

the beta access to this thing and see it

kind of in effect for real but let me

play you that clip that I was talking

about from this channel that you can see

here on this channel 6 months ago back

then I was asking how long until we have

something like this is it a year 5 years

10 years 20 and I don't know if this

thing will exactly hit all those

criteria it might but the rate of

progress here is quite astonishing from

6 months ago to to today we can't even

make babies that fast imagine this if

you can you wake up in the morning you

have your coffee or whatever other

stimulant gets you going you turn on

your computer and pops up your very own

autonomous AI agent you named him

goalgetter or GG for short that was very

clever of you Gigi reports to you what

it's been doing while you were sleeping

and it's been very busy you see you

created GG to run a business for you the

business idea is simple millions of

people in the world are looking for

simple inexpensive custom automation

Solutions some want a little script that

helps them automate their home some want

a morning routine automation where the

alarm rings their curtains open the

coffee maker turns on Etc some want

their emails answered and sorted in very

specific ways some want a thing that

feeds their cat automatically other

people want solutions for their business

that are a little bit more advanced

requires cameras and tracking Etc but

the point is that people are lazy people

are busy and you have developed quite a

reputation for delivering smart and

inexpensive custom automation Solutions

one-of-a-kind automation Solutions

people go on your website they type in

what they want done they don't even have

to know exactly what they want they just

type in something like my stupid cat

keeps running away now you respond

within a minute or two with an exact

plan of how to fix that problem and

multiple choices with different price

options

depending on how fancy they want to get

starting from a simple tracker for your

cat to an army of drones folling it

around everywhere it goes they choose

whatever option they want and they

provide a credit card for you to build

once the automation is done you type a

script doing the thing that they want

maybe you have to order some parts like

trackers or video cameras and have it

shipped to their house and then you

record a quick voice over with visuals

on how to set it all up you're kind of

like the Ikea of automations the

instructions are great they're brain

dead simple you made sure the

documentation can be understood by

anyone regardless of their Tech

background if they can use a smartphone

they can set up the thing that you've

made for them if they run into any

issues you provide unlimited support

through text emails or whatever they

want once you send everything over to

them all the instructions all the

details Etc you charge their credit card

and you're done but here's the key you

didn't do any of that you were sleeping

the whole time remember everything was

done by GG it talk to the customers and

figure out what they wanted it

brainstorm some Solutions told them

where to put their credit card

information Etc then it went to work

building needed software ordering the

parts that were needed and writing out

very clear instructions for how to use

the thing that it just built it shipped

the product either digitally or the

physical components that I needed it

handled the customer service it made

sure that everybody was happy Etc

recently you've been teaching it how to

do marketing campaigns and you're seeing

some great results of that it learned to

hug people in with little free

automations then it goes online to find

whatever information about them that it

can and then starts pitching them ideas

that it thinks they will respond to it

produces content showing how their lives

could be improved by these various

Solutions custom made tailor made to

them now your customers are ecstatic and

you are basically printing money you

don't do any work to run the business

zero the only work you do is on

improving the agent the autonomous AI

agent that is running this whole thing

you work on optimizing it on teaching it

new skills

now you are aware that eventually people

will catch on and these agents will be

more commonplace and more available to

everyone but until then you basically

found an unlimited money glitch GG so my

question for you is when will this

scenario likely play out never is it

just science fiction is it possible 10

years in the future 50 what do you think

what number of years will pass before a

handful of people have something like

this

running