16 March 2023

What’s Changing the Landscape for Business Owners: Thoughts on AI, Crypto and more with Daniel DiVenere

1Frameworks

8Insights

Listen

Frameworks in this episode

Innovation7 steps

Constraint-Driven AI Content Creation

Direct AI with precise constraints, then curate and refine its output

Read →

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Myth Buster· 1

Myth Buster12:32

AI Output Can Look Competent While Remaining Painfully Average

The speakers challenge the assumption that generated copy is automatically professional-quality copy. Daniel explains that models trained across internet-scale material tend to reproduce an average of what they ingest, while experts are more likely to recognize the resulting weaknesses.

Beginners may perceive plausible output as expert work.
Experienced practitioners notice when generated work falls below professional standards.
Training on broad internet data encourages average patterns.
A fulfilled prompt does not guarantee an effective result.

“it's probably good enough for like people that don't know too much about the subject”
Daniel DiVenere · 13:30

“they just basically swallow up the whole entire internet and then they spit out the average of what they get.”
Daniel DiVenere · 13:47

#ai quality#expertise#copywriting#content

Hot Take· 1

Hot Take10:51

Why AI Wrappers Without a Core Product May Lose Customers

Daniel distinguishes businesses built almost entirely on third-party AI functionality from established products that add AI to a broader feature set. As foundational models become widely accessible, a thin wrapper can lose its differentiation because competitors and customers gain access to the same underlying capability.

An AI feature is different from an AI-dependent business.
Thin products can be vulnerable when model APIs become broadly available.
Access to the same technology reduces feature-level differentiation.
Existing products may use AI to strengthen rather than replace their core value.

“I'd like to make the distinction between the companies that are making AI their core feature and the companies that are using AI as like,…”
Daniel DiVenere · 10:51

“the more people that have the technology, the less valuable it becomes.”
Jasmine Star · 12:10

#ai startups#saas#commoditization#product strategy

Explainer· 3

Explainer05:55

Why Daniel Views Crypto as a Four-to-Five-Year Cycle

Daniel compares the current downturn with the boom and crash that followed Bitcoin's 2017 peak. He argues that crypto repeatedly moves through periods of hype, collapse, and renewed growth rather than following a straight trajectory.

Peak enthusiasm has historically preceded steep corrections.
Bitcoin fell after its 2017 high before later reaching a much higher peak.
Daniel frames crypto as a cyclical rather than linear market.
The discussion explicitly avoids presenting investment advice.

“I think we're in this like four to five year bubble with crypto where just very cyclical.”
Daniel DiVenere · 06:31

“not investment advice”
Jasmine Star · 06:42

#bitcoin#crypto cycles#bear market#investing

Explainer15:10

Can Anyone Reliably Separate Original Work From AI Derivatives?

Training a model on a creator's archive may improve voice consistency, but Daniel warns that it can also produce material too similar to what already exists. He identifies a broader unresolved problem: distinguishing derivative AI content from genuinely original human work.

A creator's archive can help an AI imitate their voice.
Limited source material may produce repetitive results.
AI is increasingly effective at mimicking human language.
Reliable detection of derivative AI writing remains unresolved.

“the content it generates might just be too similar to the content you already have.”
Daniel DiVenere · 15:58

“How do we tell the AI generated content that's derivative from the unique original content that humans are writing?”
Daniel DiVenere · 16:26

#originality#ai detection#content creation#authorship

Explainer20:29

Prompt Marketplaces Could Become the Preset Shops of Generative AI

Daniel describes platforms where users buy and sell prompts for text and image models. He compares prompts to photography presets: reusable starting points that reveal how another creator achieved a result and can then be modified for a new use case.

Some platforms already facilitate prompt exchange.
Visual examples make image prompts especially easy to evaluate.
Purchasing a prompt provides a reusable starting configuration.
Prompt products resemble presets, templates, cheat codes, or textbooks.

“It's like an eBay for prompts”
Daniel DiVenere · 21:32

“it's a new version of like, uh, selling camera presets or like editing presets for photography.”
Daniel DiVenere · 22:29

#prompt marketplace#dall-e#digital products#ai business

Story· 1

Story01:52

Daniel Reveals His Crypto Losses—and Why He Still Believes

Daniel discloses money trapped in Voyager, losses associated with Terra Luna, and spending on NFTs that became nearly valueless. Despite those outcomes, he says his long-term belief in cryptocurrency as both an investment vehicle and a force for change remains intact.

Daniel had $20,000 locked in the failed Voyager exchange.
He recovered most of a $5,000 Terra Luna position while it crashed.
He estimates spending about $2,000 on NFTs that became nearly valueless.
The losses did not eliminate his long-term conviction in crypto.

“I have $20,000 lost in Voyager.”
Daniel DiVenere · 02:54

“I've lost that, but that still doesn't really deter me.”
Daniel DiVenere · 05:03

#crypto#voyager#terra luna#nfts#risk

Tool· 1

Tool25:22

Could DALL-E Turn a Phone Snapshot Into a Product Photo?

Daniel uses his family's coffee-and-donut business to illustrate a practical image-generation use case. Instead of professionally photographing roughly a hundred product variations, a business might submit simple phone photos and ask an image model to create studio-style versions or replacement backgrounds.

Large product catalogs make conventional photography expensive and time-consuming.
A basic real-world image can serve as input to an image model.
The model may help improve backgrounds or simulate studio presentation.
The proposed workflow was an experiment, not a proven production result.

“we're looking at like maybe a hundred images that need to be taken, right? And professionally at that.”
Daniel DiVenere · 25:23

“ask it to render a more professional version of the image.”
Daniel DiVenere · 25:40

#product photography#dall-e#small business#image generation

Takeaway· 1

Takeaway29:29

Start AI Adoption With the Work That Drains Time, Not Your Core Value

Daniel recommends experimenting with AI to identify a frequently repeated process that consumes time without representing the business's central expertise. The goal is not automation for its own sake, but freeing the owner to focus on work that directly creates value or revenue.

Begin by experimenting rather than making a large commitment.
Look for frequent and time-consuming processes.
Prioritize peripheral work over the business's core competency.
Use saved time for higher-value and revenue-generating activity.

“Maybe there's like this one process that you're doing very frequently that takes up a lot of your time that could be automated through this.”
Daniel DiVenere · 29:38

“it will just free you up to do other things in your important things that actually bring you revenue.”
Daniel DiVenere · 30:07

#automation#small business#experimentation#productivity