HomeVideos

This New AI Agent Is Basically an Entire Company

Now Playing

This New AI Agent Is Basically an Entire Company

Transcript

355 segments

0:00

This business used to be my entire life,

0:03

but I lost my passion for it and I

0:06

abandoned it for years. But then my X

0:09

account got banned and overnight my

0:11

business lost its main source of

0:13

traffic. So I needed to replace that

0:16

income fast. So I'm going to use AI to

0:20

revive my old business. [music] Building

0:22

it the first time took me years of

0:24

grinding, but I don't have years. So

0:27

[music] AI is going to need to build

0:29

this for me and learn how to go viral on

0:32

social media. Create and post content,

0:35

generate leads and close those leads for

0:37

me and run the whole business for me

0:40

start to finish. Okay, so the first

0:43

thing I needed to do was build my AI

0:45

dream team. But I needed an AI tool that

0:48

can reliably create entire agent teams

0:52

that I could hook different models in

0:54

and that I could connect to all of my

0:56

different tools in order to revive my

0:58

business as fast as possible. So I got

1:02

to researching and found this very cool

1:04

product called HyperAgent. The reason I

1:07

decided to use it for this test is

1:09

simple. I do not just need a chatbot. I

1:13

quit this page because it got repetitive

1:15

sourcing content, writing captions,

1:18

scheduling posts, replying to DMs every

1:21

single week. But AI doesn't get bored. I

1:24

also reached out to the team and they

1:26

love this idea so much that they were

1:28

very happy to sponsor today's video. big

1:32

shout out to the HyperAgent team. This

1:34

video literally wouldn't be possible

1:36

without them. HyperAgent is giving away

1:38

$100 in bonus credits by the way when

1:41

you sign up. So check it out now inside

1:43

the link in the description below. And

1:45

with that I got started and built five

1:49

agents. So I'm building out my team and

1:51

I have to say I really enjoy the

1:53

platform. I really enjoy how it looks

1:56

like, how it feels, even hiring the

1:58

agents one by one. So, we currently have

2:01

our team, an auditor, a researcher, a

2:03

strategist, a content creator, and a

2:05

salesperson, and we're still hiring one

2:09

more, which is the operator, right now.

2:11

And it's done a pretty good job at

2:13

understanding the page so far. I'm

2:16

giving it as much information as it

2:17

needs, and from there we'll get started

2:20

with day one. All right, and we're

2:23

wrapping up day one. It just posted its

2:26

first post literally 3 minutes ago. We

2:28

already have six likes, which is great.

2:30

And this is the first post that it

2:32

helped us spin up and make. This is the

2:35

operator agent's daily job, actually

2:38

writing out each post for the day with

2:40

their captions, media over here,

2:43

and the time slots it scheduled it out

2:46

for for the day. And yeah, we have two

2:49

posts scheduled per day. Day one, I'm

2:51

focusing on the content itself. I'm

2:54

going to let today's post run and come

2:56

back and check tomorrow. But, a couple

2:58

of things I've noticed, it's been 3

3:00

years almost since I've posted.

3:02

Literally, the post before it was 151

3:05

weeks ago. That's almost 3 years, if not

3:07

more. A lot has changed on Instagram.

3:09

The researcher has come back and was

3:11

literally like, "Dude, like I don't

3:13

think we can continue posting the things

3:15

you used to post. Policies have changed.

3:17

The Instagram has changed. The internet

3:20

has changed since then, and it wouldn't

3:22

be fair to post these today." Uh so, we

3:25

need to rework new strategy, something

3:27

I've never tried before. It will have to

3:29

come up with it, which makes this

3:31

challenge even more interesting. That's

3:33

the researcher's report. It pulled

3:35

Instagram's current policy in 2026 and

3:39

told me flat-out that the old format is

3:41

a ban risk. And that right here at the

3:44

bottom is the strategist coming up with

3:47

completely new content plan. See you in

3:49

day two. Day two here and I'm genuinely

3:53

impressed with what we've been able to

3:56

do in day one and so far. I mean, the

3:59

first post we've posted has done very

4:02

very well. Its reception was taken very

4:04

well. Just evidenced by this comment,

4:06

"Well, finally the admin is back." I do

4:09

suspect though that, as you can see, the

4:11

post after it did a little worse and we

4:14

just posted another post right now on

4:16

day two. I suspect that this was because

4:19

this was my first post in a long time.

4:21

Instagram notifies a lot of people, you

4:23

know, like, "Hey, this person has posted

4:24

for the first time in a while." I

4:26

suspect that might be why, but what I'm

4:29

more so surprised by is Hyper Agent's

4:33

ability to just work so well with a

4:35

fleet of agents. I haven't been talking

4:39

to it that much, but it's been able to

4:41

do a lot of research where needed. It's

4:43

been very spot-on, very accurate. I'm

4:46

going to be testing new angles, like

4:49

news for example, on that page. And it's

4:51

crazy that it's been able to do all of

4:53

that on its own. And what's even

4:56

crazier, something that I failed to

4:58

realize when I first started this

4:59

challenge, I thought it would be a lot

5:02

closer to posting the same type of

5:04

content I used to post before. But

5:07

thanks to Hyper Agent, it's literally

5:09

gone and done a deep dive on the state

5:12

of Instagram right now. And I didn't

5:14

even know this, but I can't post the

5:17

same type of content I used to post

5:19

anymore because Instagram has apparently

5:21

flagged that as a type of content that

5:24

if you post, you can either get banned

5:26

or you can just, you know, be

5:28

shadowbanned. Your post won't show up to

5:30

anyone, including your followers, which

5:33

we don't want that to happen. So, it

5:35

literally spun up a strategy that I

5:37

wouldn't even have thought of or had the

5:40

the time or patience to come up with,

5:42

and it's currently testing it out. This

5:44

is insane. I'm pleasantly surprised with

5:46

this. It's day four and honestly, I'm

5:49

surprised. Yesterday's post took off

5:53

absolutely that well. It did even better

5:55

than the first post that we posted

5:58

almost two times as good. Even though it

6:00

didn't get that many comments, I think

6:02

news are great thing we have to play

6:05

into for the rest of this video and it's

6:08

such a great strategy. Honestly, I've

6:11

told it to double down. We just posted

6:12

two. So, we'll see how today's posts go.

6:15

But I love how it's just learning. It's

6:18

literally taking up every single post

6:20

trying to see what does well, what

6:22

doesn't do well, monitoring competitors,

6:24

seeing how to go viral. I wouldn't have

6:27

figured this out on my own to be

6:29

completely honest or it would have taken

6:31

me months if not years to figure this

6:34

out and I would have either gotten

6:35

banned or shadow banned at that point.

6:38

The second thing I've done today so far

6:41

is start the DM replying which was just

6:46

such a headache back then to have to

6:49

find, reply, answer all the questions.

6:52

The great thing with AI is you can

6:54

literally just give it that

6:56

responsibility and it will do a 10 times

6:58

better job than you ever could. What's

7:01

nice is I can talk to specific agents

7:03

directly. This is me telling the sales

7:06

agent, "Hey, we need to start DMs and

7:08

selling. How do we get started?" It

7:10

asked me a list of questions that it

7:12

needs and then we literally started. I

7:14

set up an automation so that it sends me

7:16

an email. It replies to DMs and it will

7:18

stop before pitching a price. Oh, I'm

7:21

excited. I'll be on the lookout to see

7:24

how we do by day five. Day five and I'm

7:28

pleasantly surprised with the results

7:31

that we've had today. I mean, it decided

7:33

to double down on news which I mean,

7:36

absolutely because just look at this

7:38

post right here. We be about 11 hours

7:41

ago so in the morning, and it's

7:43

absolutely blowing up. Uh we can see the

7:46

reach on it, but on the phone it shows

7:48

that it's currently at 80,000 people.

7:51

It's crazy that it understood how to go

7:55

viral. I mean, this post right here is

7:57

doing better than posts I had around the

8:00

peak of this page's lifespan. So, this

8:03

is absolutely wild. This is update

8:05

number one. Update number two. It did

8:07

start replying to a bunch of DMs, all

8:10

the DMs we're getting, and there's no

8:14

lead yet that is looking like a

8:16

promising one, or that is looking like

8:19

they're about to pay. But, it's going to

8:21

keep going, and the great thing is we

8:23

have it up and running, and yeah, that's

8:26

literally it. I can't wait to see what

8:28

we do on day six. Day six, and oh my

8:31

god, guys, yesterday's post, let me just

8:34

refresh real quick, went absolutely

8:37

viral. I mean, look at that. 3,000

8:40

likes, over 150,000

8:43

impressions on this just one post alone.

8:45

We had other posts do well as well.

8:48

Today was a very good day in terms of

8:51

posts, and yeah, the best part is we got

8:54

to enjoy the lake. We went, we sailed,

8:58

we had fun, and I didn't have to worry

9:00

about any of this. I mean, typically it

9:02

was on Sundays, days like this, where I

9:05

would spend the day sourcing content,

9:08

creating the content, just hours of my

9:10

day spent on creating that content, and

9:14

I just got to rest, to relax, to enjoy,

9:16

to make good quality memories, which

9:19

makes me so, so happy. On the downside,

9:23

though, the DMs don't look that

9:25

promising.

9:26

>> I mean, you could see it in the pipeline

9:28

with all of these statuses, all of these

9:30

follow-ups sent, not a single reply

9:32

received.

9:33

10 active leads at the moment, zero

9:36

replies, zero objections, zero purchase

9:39

intent.

9:40

Oh god. I mean, it's been replying to

9:42

every single DM we've received and it's

9:45

gone as far as literally one guy just

9:48

deactivated this account. This is how

9:50

bad the DMs are going. I don't know if

9:52

we're going to make a sale this week.

9:54

I'm not very hopeful. We have only one

9:57

day left to this challenge, but only

9:59

time will tell. I mean, one thing I know

10:02

is even if we don't make sales this

10:04

week, I know it's going to happen this

10:07

month. I know we need to get the ball

10:09

rolling first. I mean, content is number

10:12

one. If your content sucks, your reach

10:14

sucks. If your reach sucks, your

10:16

discovery sucks. And if your discovery

10:18

sucks, then your pipeline doesn't exist.

10:20

So, content is number one and it seems

10:23

to really understand what makes good

10:26

content. I mean, we just posted this and

10:27

it's already doing well. See you in the

10:30

final day of this challenge. Oh my god,

10:32

I was getting ready to wrap up and

10:35

writing down the results for this

10:37

challenge on day seven and we just got

10:41

our first close. Finally. I honestly

10:45

didn't think that we would get a single

10:48

close during the 7-day challenge. I

10:50

mean, 7 days is insanely short to revive

10:53

an entire business. I'm shook right now.

10:56

But I mean, on a positive note, look at

10:58

this. We took a page that was completely

11:01

dead for three entire years and we

11:04

managed to bring it back to life. So,

11:08

after 1 week, here is what happened. The

11:12

page started with no posts whatsoever

11:14

for the past 3 years. However, in the

11:17

last 7 days, we've managed to post 20

11:21

content pieces this week, generate

11:23

668,000

11:26

impressions, reach 151,000

11:29

unique accounts, and receive 13,900

11:34

interactions or engagement. We did lose

11:37

242 followers in total, but that was

11:41

mostly because on day one of this

11:42

challenge, hundreds of people unfollowed

11:45

as soon as we got active again. I mean,

11:47

you can see that massive dip on our

11:50

first day of this challenge, but then

11:53

you can see we're now above break even.

11:55

We're now finally starting to gain

11:57

followers again. And that is classic for

12:00

a business that's been dormant for 3

12:02

years. We did get 18 total inbound leads

12:05

and closed one, generating a total of

12:09

$50 in revenue on week one, which isn't

12:12

a massive life-changing number, but it's

12:15

honestly more than I ever expected on

12:18

week one. But, I don't think that is the

12:21

only useful way to judge this. A dormant

12:24

page does not usually go from dead to

12:26

profitable in 7 days. The better

12:29

question is whether the system created

12:31

leading indicators that make the page

12:33

worth continuing. I mean, we did have

12:36

one post reach 179,000

12:39

views at the time of this recording,

12:41

with many others getting around 50 to

12:44

100,000 views. I am pleasantly surprised

12:47

by how well my AI agents were able to

12:50

handle autonomy and pump out much better

12:53

content than I would have ever been able

12:56

to. So, the big question is, after 7

12:58

days of trying this out, would I keep

13:01

this running? And the answer is a big

13:03

fat giant yes, absolutely. We managed to

13:07

[clears throat] turn a dead business

13:09

asset into a working pipeline again. I

13:12

mean, we got a strategy, we created

13:14

content, we replied to leads, and this

13:16

AI agent gave me a reason to keep

13:19

operating the page instead of letting it

13:21

just sit there. And that is a massive

13:24

win. If you want to try this out

13:26

yourself, I will leave a link for you to

13:29

try out HyperAgent in the description

13:31

below. You can get $100 worth of credits

13:34

when you sign up. Look, here's my advice

13:36

after what I've learned reviving my dead

13:39

business. Do not start with a giant

13:41

company-wide automation. Start with one

13:44

agent that does one useful job every

13:46

morning. Maybe it researches leads.

13:48

Maybe it prepares a content brief. Maybe

13:50

it checks your pipeline. One single

13:53

agent, one single job. That's literally

13:56

the whole starting point. And once that

13:59

agent creates a clean handoff, add the

14:02

second agent. That is when it starts

14:04

feeling less like a chatbot and more

14:06

like a team. That is the difference

14:08

between AI that gives you a reply and AI

14:11

that prepares work when you actually use

14:13

it. So, did AI revive my dead business?

14:17

I would say yes. My wife is actually the

14:20

one who suggested this whole video and

14:22

who asked me if AI could bring this

14:25

thing back to life. And I told her

14:28

probably not, but it took something I

14:30

was ignoring for years and it helped me

14:33

automate the parts that made me lose

14:35

passion for it. And that is usually

14:37

where businesses die in the first place.

14:39

Not because the opportunity is gone, but

14:41

because nobody keeps pushing the boring

14:44

parts forward. Comment HyperAgent if you

14:47

enjoyed this type of challenge video.

14:49

I'd love to know if there is interest

14:51

for me to do more of these. And if you

14:54

enjoyed this video, make sure to leave a

14:56

like and subscribe if you're new to my

14:58

channel because I have a ton more

15:00

content coming your way. Anyway, I'll

15:02

see you in the next video very, very

15:05

soon. See you later.

Interactive Summary

The creator attempts to revive a dormant business that had been inactive for three years by using an AI platform called HyperAgent. Over the course of seven days, he deploys a team of AI agents to handle content creation, strategy, and sales lead management. The experiment results in significant growth in social media reach and engagement, eventually leading to a successful sale, proving that AI can effectively manage repetitive tasks and automate business operations.

Suggested questions

4 ready-made prompts