StarZero All shows Try the podcast clipper →

Mar 26, 2025 · 41m · startup-ideas

I challenged Manus AI to find $1M+ AI Startup Ideas - It Actually Delivered

Greg Isenberg · 34m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Entrepreneur Greg Isenberg stress-tests the autonomous agent Manus AI to research, validate, and build a viable million-dollar startup concept from scratch. Through live experimentation, he demonstrates how autonomous multi-agent workflows can execute complex market research and code generation when paired with active human guidance.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 100% of the talking time here. How this is scored →

Greg as informed peer 0.0 Guest teaching 0.0 Guest disagreement 0.0 Greg pushing back 0.0
05100:0015:0030:001:27–5:22 · Greg as informed peer 0/10 Formulating the Prompt and Launching Manus's Autonomous Agent This is a solo monologue segment where Greg sets up the prompt for Manus AI. Because there is no human guest dialogue, all interaction scores are 0.5:22–12:15 · Greg as informed peer 0/10 Reviewing Manus AI Capabilities and Practical Use Cases Greg walks through sample use cases and features of Manus AI while waiting for the business plan output. As a solo demonstration monologue, all scores remain 0.12:15–16:44 · Greg as informed peer 0/10 Analyzing Initial YC Findings and Guiding Reddit Research Greg examines the initial findings returned by Manus AI and guides its Reddit research parameters. Being a solo monologue, host and guest interaction scores are 0.16:44–22:24 · Greg as informed peer 0/10 Evaluating Niche Startup Ideas and Reddit Insights Greg analyzes the generated niche AI startup concepts. A brief transcription artifact occurs with SPEAKER_01 uttering a single word, but the segment remains a solo monologue with all scores at 0.22:24–27:35 · Greg as informed peer 0/10 Demanding Concrete Search Data and Validating Demand Greg prompts the AI to deliver empirical Google Trends and subreddit subscriber data rather than broad statistics. As a monologue demonstration, all interaction scores are 0.27:35–33:22 · Greg as informed peer 0/10 Generating the SmartConsult AI MVP and Landing Page Greg observes Manus coding a full landing page and MVP strategy via terminal and browser automation. As a solo walkthrough monologue, all scores are 0.33:22–38:22 · Greg as informed peer 0/10 Reviewing the Executive Summary and Testing the Live Landing Page Greg downloads, reviews, and tests the generated HTML landing page and gives final takeaway thoughts on managing autonomous AI agents. All scores are 0 for monologue format.1:27–5:22 · Guest teaching 0/10 Formulating the Prompt and Launching Manus's Autonomous Agent This is a solo monologue segment where Greg sets up the prompt for Manus AI. Because there is no human guest dialogue, all interaction scores are 0.5:22–12:15 · Guest teaching 0/10 Reviewing Manus AI Capabilities and Practical Use Cases Greg walks through sample use cases and features of Manus AI while waiting for the business plan output. As a solo demonstration monologue, all scores remain 0.12:15–16:44 · Guest teaching 0/10 Analyzing Initial YC Findings and Guiding Reddit Research Greg examines the initial findings returned by Manus AI and guides its Reddit research parameters. Being a solo monologue, host and guest interaction scores are 0.16:44–22:24 · Guest teaching 0/10 Evaluating Niche Startup Ideas and Reddit Insights Greg analyzes the generated niche AI startup concepts. A brief transcription artifact occurs with SPEAKER_01 uttering a single word, but the segment remains a solo monologue with all scores at 0.22:24–27:35 · Guest teaching 0/10 Demanding Concrete Search Data and Validating Demand Greg prompts the AI to deliver empirical Google Trends and subreddit subscriber data rather than broad statistics. As a monologue demonstration, all interaction scores are 0.27:35–33:22 · Guest teaching 0/10 Generating the SmartConsult AI MVP and Landing Page Greg observes Manus coding a full landing page and MVP strategy via terminal and browser automation. As a solo walkthrough monologue, all scores are 0.33:22–38:22 · Guest teaching 0/10 Reviewing the Executive Summary and Testing the Live Landing Page Greg downloads, reviews, and tests the generated HTML landing page and gives final takeaway thoughts on managing autonomous AI agents. All scores are 0 for monologue format.1:27–5:22 · Guest disagreement 0/10 Formulating the Prompt and Launching Manus's Autonomous Agent This is a solo monologue segment where Greg sets up the prompt for Manus AI. Because there is no human guest dialogue, all interaction scores are 0.5:22–12:15 · Guest disagreement 0/10 Reviewing Manus AI Capabilities and Practical Use Cases Greg walks through sample use cases and features of Manus AI while waiting for the business plan output. As a solo demonstration monologue, all scores remain 0.12:15–16:44 · Guest disagreement 0/10 Analyzing Initial YC Findings and Guiding Reddit Research Greg examines the initial findings returned by Manus AI and guides its Reddit research parameters. Being a solo monologue, host and guest interaction scores are 0.16:44–22:24 · Guest disagreement 0/10 Evaluating Niche Startup Ideas and Reddit Insights Greg analyzes the generated niche AI startup concepts. A brief transcription artifact occurs with SPEAKER_01 uttering a single word, but the segment remains a solo monologue with all scores at 0.22:24–27:35 · Guest disagreement 0/10 Demanding Concrete Search Data and Validating Demand Greg prompts the AI to deliver empirical Google Trends and subreddit subscriber data rather than broad statistics. As a monologue demonstration, all interaction scores are 0.27:35–33:22 · Guest disagreement 0/10 Generating the SmartConsult AI MVP and Landing Page Greg observes Manus coding a full landing page and MVP strategy via terminal and browser automation. As a solo walkthrough monologue, all scores are 0.33:22–38:22 · Guest disagreement 0/10 Reviewing the Executive Summary and Testing the Live Landing Page Greg downloads, reviews, and tests the generated HTML landing page and gives final takeaway thoughts on managing autonomous AI agents. All scores are 0 for monologue format.1:27–5:22 · Greg pushing back 0/10 Formulating the Prompt and Launching Manus's Autonomous Agent This is a solo monologue segment where Greg sets up the prompt for Manus AI. Because there is no human guest dialogue, all interaction scores are 0.5:22–12:15 · Greg pushing back 0/10 Reviewing Manus AI Capabilities and Practical Use Cases Greg walks through sample use cases and features of Manus AI while waiting for the business plan output. As a solo demonstration monologue, all scores remain 0.12:15–16:44 · Greg pushing back 0/10 Analyzing Initial YC Findings and Guiding Reddit Research Greg examines the initial findings returned by Manus AI and guides its Reddit research parameters. Being a solo monologue, host and guest interaction scores are 0.16:44–22:24 · Greg pushing back 0/10 Evaluating Niche Startup Ideas and Reddit Insights Greg analyzes the generated niche AI startup concepts. A brief transcription artifact occurs with SPEAKER_01 uttering a single word, but the segment remains a solo monologue with all scores at 0.22:24–27:35 · Greg pushing back 0/10 Demanding Concrete Search Data and Validating Demand Greg prompts the AI to deliver empirical Google Trends and subreddit subscriber data rather than broad statistics. As a monologue demonstration, all interaction scores are 0.27:35–33:22 · Greg pushing back 0/10 Generating the SmartConsult AI MVP and Landing Page Greg observes Manus coding a full landing page and MVP strategy via terminal and browser automation. As a solo walkthrough monologue, all scores are 0.33:22–38:22 · Greg pushing back 0/10 Reviewing the Executive Summary and Testing the Live Landing Page Greg downloads, reviews, and tests the generated HTML landing page and gives final takeaway thoughts on managing autonomous AI agents. All scores are 0 for monologue format.

speaking balance: gold is Greg, purple is the guest (3 minute bins)

0:00 · Greg 100% · guest 0%0:00 · Greg 100% · guest 0%3:00 · Greg 100% · guest 0%3:00 · Greg 100% · guest 0%6:00 · Greg 100% · guest 0%6:00 · Greg 100% · guest 0%9:00 · Greg 100% · guest 0%9:00 · Greg 100% · guest 0%12:00 · Greg 100% · guest 0%12:00 · Greg 100% · guest 0%15:00 · Greg 100% · guest 0%15:00 · Greg 100% · guest 0%18:00 · Greg 100% · guest 0%18:00 · Greg 100% · guest 0%21:00 · Greg 99.9% · guest 0.1%21:00 · Greg 99.9% · guest 0.1%24:00 · Greg 100% · guest 0%24:00 · Greg 100% · guest 0%27:00 · Greg 100% · guest 0%27:00 · Greg 100% · guest 0%30:00 · Greg 100% · guest 0%30:00 · Greg 100% · guest 0%33:00 · Greg 100% · guest 0%33:00 · Greg 100% · guest 0%36:00 · Greg 100% · guest 0%36:00 · Greg 100% · guest 0%39:00 · Greg 100% · guest 0%39:00 · Greg 100% · guest 0%
Sharpest disagreement ▶ 21:45 Zero guest combativeness in solo demo

This is a solo tool-testing monologue without an active interview guest to exhibit pushback or combativeness.

 Hardest push from Greg ▶ 22:24 Host prompting challenge to AI tool

Greg pushes back against high-level macro statistics by explicitly demanding raw Google Trends and subreddit growth data from the AI agent.

 Biggest teaching moment ▶ 12:15 Zero schooling in solo walkthrough

There is no human guest present to educate or correct the host during this software test.

 Greg holds their own ▶ 34:35 Host startup framework commentary

Greg draws on his background selling venture-backed businesses to critique AI outputs and discuss managing AI agents like junior analysts.
the scores for every segment, with the reasoning behind each
ChapterTopicGreg as informed peerGuest teachingGuest disagreementGreg pushing backWhy
Formulating the Prompt and Launching Manus's Autonomous Agent 0000 This is a solo monologue segment where Greg sets up the prompt for Manus AI. Because there is no human guest dialogue, all interaction scores are 0.
Reviewing Manus AI Capabilities and Practical Use Cases 0000 Greg walks through sample use cases and features of Manus AI while waiting for the business plan output. As a solo demonstration monologue, all scores remain 0.
Analyzing Initial YC Findings and Guiding Reddit Research 0000 Greg examines the initial findings returned by Manus AI and guides its Reddit research parameters. Being a solo monologue, host and guest interaction scores are 0.
Evaluating Niche Startup Ideas and Reddit Insights 0000 Greg analyzes the generated niche AI startup concepts. A brief transcription artifact occurs with SPEAKER_01 uttering a single word, but the segment remains a solo monologue with all scores at 0.
Demanding Concrete Search Data and Validating Demand 0000 Greg prompts the AI to deliver empirical Google Trends and subreddit subscriber data rather than broad statistics. As a monologue demonstration, all interaction scores are 0.
Generating the SmartConsult AI MVP and Landing Page 0000 Greg observes Manus coding a full landing page and MVP strategy via terminal and browser automation. As a solo walkthrough monologue, all scores are 0.
Reviewing the Executive Summary and Testing the Live Landing Page 0000 Greg downloads, reviews, and tests the generated HTML landing page and gives final takeaway thoughts on managing autonomous AI agents. All scores are 0 for monologue format.

Statements from this episode (9)

Assertion Supported
Isenberg: Manus AI lets users manually control its agent browser sessions
“There is a button that if you wanted, you can take control of Madness, and you can actually use Madness's computer.”
Greg Isenberg Mar 26, 2025 ▶ 4:49
Assertion Not checkable as stated
Isenberg: Complex research prompts in Manus AI take up to 15 minutes
“So what I've noticed about Madness is, if you want to do a prompt like this, it's going to take 10 to 15 minutes. So just so you know, like, it's not going to take a minute or 30 seconds. Like, this generally takes 10 to 15 minutes.”
Greg Isenberg Mar 26, 2025 ▶ 4:58
Opinion
Isenberg: AI agents like Manus can replace six-figure investment banking analysts
“You know, at JP Morgan, you're paying people hundreds of thousands of dollars to do this, and look, you know, look what's happening.”
Greg Isenberg Mar 26, 2025 ▶ 7:47
Opinion
Isenberg: Manus AI feels more powerful than ChatGPT, Perplexity, and Claude
“What are you noticing about Manus is it feels like just a little more horsepower than chat GPT, a little more horsepower than perplexity, a little more horsepower than Claude. So it's because of this like multi-agent browsing sort of capabilities, I think”
Greg Isenberg Mar 26, 2025 ▶ 9:40
Insight
Isenberg: Manus AI cannot build a $1M business without precise human prompting
“You can't rely on Manus to basically say, hey, build me a million dollar business and just like say, go wild. You do have to do the right prompts here.”
Greg Isenberg Mar 26, 2025 ▶ 16:31
Insight
Isenberg: AI agents can replace expensive specialized teams tracking regulatory changes
“What used to cost a lot of money to go in you know, hire a team, specialized team who understands all these regulatory changes. Now you can just use something like Manus to actually go and figure all these stuff, this stuff out.”
Greg Isenberg Mar 26, 2025 ▶ 19:10
Opinion
Isenberg: Niche AI aggregation for regulated sectors is the best startup idea
“I actually think the most interesting one is niche specific AI information aggregation for regulated industries.”
Greg Isenberg Mar 26, 2025 ▶ 19:59
Prediction Open · timeframe Mar 2030
Isenberg predicts a $1M 'Design Pickle for AI agents' will emerge
“And I assume that that's going to happen in, in this space as well. Like you'll be, someone is going to create the design pickle, but for, you know, AI and agents. So if you can do that there's a path to a million dollars. That's for sure. Maybe even more.”
Greg Isenberg Mar 26, 2025 ▶ 25:02
Insight
Isenberg: Current AI agents act like junior employees requiring active human management
“It does, it really does feel like an AI employee, but it's kind of like when you have a, you know, junior-ish employee you do have to manage them, mentor them, and get them to be smart.”
Greg Isenberg Mar 26, 2025 ▶ 39:43
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more