Recent Posts

105 High-Value Business Niches

Written by Ricky Nave in Markets,Media & Content Businesses

Whether you want to start a business, build an audience on social media, start a blog, or create a marketing campaign, your first task is exactly the same: choose a “niche”.

A niche is just a group of people or businesses who share a set of common interests or problems. More specifically, your niche is your set of target customers, and your niche’s common problems are what you are going to talk about and sell solutions to solve.

When choosing a niche, you want something that is both large but also sufficiently homogeneous. For example, the niche consisting of “all adults living in California” is a relatively large niche (over 30 million adults who collectively brought in over $3 trillion of personal income in 2022). However, there is essentially nothing that all these people have in common beyond the fact that they all live in California. You might be able to create a YouTube channel about general news in California (California laws & elections, weather, etc) but that’s about it.

Now let’s contrast the niche of all people in California with the niche of all small businesses in America. According to the U.S. chamber of commerce, there are over 33 million small businesses in America as of April 2023. In 2022, those businesses contributed over $11 trillion to U.S. GDP, and every single one of those businesses has an ongoing LEGAL AND PRACTICAL REQUIREMENT for bookkeeping & tax filing. That means the niche of bookkeeping & taxes for small businesses in America offers much more opportunity than the niche of general news for Californians. That’s why Quickbooks has over 26 million users and generates billions of dollars in revenue each year. It’s also why there are dozens of podcasts talking about how to build a bookkeeping business.

Rule of Thumb: Small businesses are the low-hanging fruit niches. They are easier to sell to than consumers, take less time to sell to than large businesses, and have money to spend on anything that can help them make more money or lose less money.

That doesn’t necessarily mean you should target ALL small businesses like Quickbooks does. Instead, you’ll find it easiest to get early customers if you find a more specific (but still large-ish) subset of small businesses to focus on (a niche within a niche).

In the table below, I list 105 examples of high-value small business niches. Each niche listed is a multi-billion dollar niche yet is homogeneous enough that all businesses in the niche share some common problems. However, if you want to narrow your niche even further (called “niching down”) you can intersect any of the niches below with a geographic area (e.g. instead of targeting “pharmacies” you might only target “pharmacies in New York State” or even “pharmacies in Manhattan”).

Alternatively, you can niche down from any of the examples below by focusing on businesses within the niche which share even more commonality. For example, instead of targeting all veterinarian businesses, you could target exotic pet veterinarian businesses.

Business TypeIndustryNumber of Unique Companies in the U.S.Total Annual Payroll
1Veterinarian businesses

NAICS 541940		Specialized Professional Services27,943$16 Billion
2Pharmacies and drug stores

NAICS 446110		Retail19,486$27 Billion
3Drugs and druggists’ sundries merchant wholesalers

NAICS 424210		Wholesale7,207$39 Billion
4Other chemical and allied products merchant wholesalers

NAICS 424690		Wholesale5,804$11 Billion
5Physicians’ offices (excluding mental health specialists)

NAICS 621111		Healthcare157,663$235 Billion
6Dentists’ offices

NAICS 621210		Healthcare124,384$50 Billion
7Physical, occupational, and speech therapist companies

NAICS 621340		Healthcare28,366$18 Billion
8Outpatient mental health & substance abuse centers

NAICS 621420		Healthcare7,215$12 Billion
9“All other outpatient care centers”

NAICS 621498

Excludes:
– Family planning centers
– Outpatient mental health & substance abuse centers
– HMO medical centers
– kidney dialysis centers
– freestanding ambulatory surgical and emergency centers

Includes (non-exhaustive list of examples):
– Outpatient biofeedback centers
– Outpatient pain therapy centers
– Outpatient community health centers
– Outpatient sleep disorder clinics
– Outpatient infusion therapy centers
– Freestanding birth centers
– Centers/clinics with health practitioners from more than one industry practicing		Healthcare6,258$28 Billion
10Medical, dental, and hospital equipment and supplies merchant wholesalers

NAICS 423450		Wholesale8,156$29 Billion
11Law firms

NAICS 541110		Specialized Professional Services161,132$110 Billion
12Insurance agencies and brokerages

NAICS 524210		Insurance122,198$54 Billion
13Real estate agencies & brokerages

NAICS 531210		Real Estate120,942$24 Billion
14Residential remodeling companies

NAICS 236118		Real Estate114,459$18 Billion
15Landscaping companies

NAICS 561730		Maintenance Services105,954$30 Billion
16Janitorial service companies

NAICS 561720		Maintenance Services58,011$25 Billion
17Family clothing stores

NAICS 448140		Retail6,710$13 Billion
18Stand alone convenience stores

NAICS 445120		Retail27,998$2.7 Billion
19Gasoline stations with convenience stores

NAICS 447110		Retail56,460$18 Billion
20Supermarkets and other grocery (except convenience) stores

NAICS 445110		Retail38,753$65 Billion
21Grocery stores

NAICS 4451		Retail66,709$68 Billion
22Health & personal care stores

NAICS 4461		Retail43,613$37 Billion
23Industrial machinery and equipment merchant wholesalers

NAICS 423830		Wholesale21,231$30 Billion
24Computer and computer peripheral equipment and software merchant wholesalers

NAICS 423430		Wholesale5,517$20 Billion
25Other electronic parts and equipment merchant wholesalers

NAICS 423690		Wholesale8,474$34 Billion
26Hardware stores

NAICS 444130		Retail9,476$4.1 Billion
27Other building material dealers

NAICS 444190

Includes:
– lumber stores
– glass stores
– fencing stores
– plumbing supplies stores
– electrical supplies stores
– kitchen and bath cabinet & countertop stores		Retail21,876$14 Billion
28Couriers and express delivery service companies

NAICS 492110		Mail & Freight Transportation4,414$32 Billion
29Freight transportation arrangement companies

NAICS 488510		Mail & Freight Transportation14,999$18 Billion
30Long-distance general freight trucking

NAICS 484121		Mail & Freight Transportation38,733$27 Billion
31Long-distance specialized freight (except used goods) trucking

NAICS 484230		Mail & Freight Transportation9,609$10 Billion
32Local general freight trucking

NAICS 484110		Mail & Freight Transportation32,883$10 Billion
33Local specialized freight (except used goods) trucking

NAICS 484220		Mail & Freight Transportation29,980$12 Billion
34Freight trucking (total)

NAICS 4841 & 4842		Mail & Freight Transportation123,378$82 Billion
35Office administrative services companies

NAICS 561110		Business & Professional Services31,298$36 Billion
36Administrative management and general management consulting companies

NAICS 541611		Consulting & Business Services80,226$70 Billion
37Construction companies focused on power and communication lines & related structures

NAICS 237130		Construction5,329$19 Billion
38“Other” scientific and technical consulting businesses

NAICS 541690		Consulting & Business Services25,255$13 Billion
39“All Other” professional, scientific, and technical services businessesConsulting & Business Services19,397$8.2 Billion
40Marketing consulting service companies

NAICS 541219		Consulting & Business Services40,946$17 Billion
41Advertising agencies

NAICS 541810		Consulting & Business Services12,545$18 Billion
42PR agencies

NAICS 541820		Consulting & Business Services7,917$5.9 Billion
43Investment advisers

NAICS 523930		Finance20,213$14 Billion
44Portfolio management companies

NAICS 523920		Finance24,128$89 Billion
45Offices of other holding companies

NAICS 551112

*These are typically corporations or LLCs that hold influential security interests of other companies but don’t actively participate in those other companies. Private equity fund management companies would would be the typical example.		Finance6,019$11 Billion
46Miscellaneous financial contract intermediation companies

NAICS 523910

These are businesses that primarily buy and sell financial contracts for their own account (typically on a spread basis), and does not include investment banks, securities dealers, or commodity contract dealers. Typical examples of businesses in this category are:
– Investment clubs
– Tax lien dealing companies (i.e. they act as a principal in dealing tax liens to investors)
– Mineral royalties or leases dealers
– Proprietary venture capital investment companies (i.e. not managing a fund)
– Mortgage dealers
– Prop trading companies
– Viatical settlement companies		Finance8,029$10 Billion
47Nondepository credit intermediation companies

NAICS 5222

These include credit card issuing companies, sales financing companies, and other nondepository credit intermediation companies		Finance14,067$59 Billion
48Activities related to credit intermediation

NAICS 5223		Finance13,761$31 Billion
49Insurance carriers

NAICS 5241		Insurance5,162$150 Billion
50Commercial banks

NAICS 522110		Finance4,473$138 Billion
51Real estate credit companies

NAICS 522292		Finance3,400$28 Billion
52Financial transaction processing, reserving, and clearinghouse businesses

NAICS 522320		Finance3,163$21 Billion
53Third party administrators of insurance and pension funds

NAICS 524292		Finance2,626$26 Billion
54Tax preparation companies

NAICS 541213		Finance17,302$2.3 Billion
55CPA firms

NAICS 541211		Finance52,111$44 Billion
56Other accounting service companies

NAICS 541219		Finance41,081$11 Billion
57Accounting, tax prep, bookkeeping, and payroll services companiesFinance114,687$74 Billion
58Residential building construction

NAICS 2361		Construction184,122$44 Billion
59Electrical contractors and other wiring installation contractors

NAICS 238210		Construction74,649$59 Billion
60Drywall and insulation contractor companies

NAICS 238310		Construction18,864$15 Billion
61Site preparation contractor companies

NAICS 238910		Construction37,102$27 Billion
62Poured concrete foundation and structure contractor companies

NAICS 238110		Construction20,947$14 Billion
63General warehousing and storage companies

NAICS 493110		Real Estate6,034$41 Billion
64Residential landlord companies

NAICS 531110		Real Estate52,216$15 Billion
65Crude petroleum extraction companies

NAICS 211120		Mining & Extraction4,250$11 Billion
66Coal mining companies

NAICS 21211		Mining & Extraction362$4.3 Billion
67Gold mining companies

NAICS 212221		Mining & Extraction141$1.5 Billion
68Crushed and broken limestone mining & quarrying companies

NAICS 212312		Mining & Extraction539$2.3 Billion
69Sand, gravel, clay, and ceramic and refractory mineral mining companies

NAICS 21232		Mining & Extraction1,534$2.3 Billion
70Companies that provide support activities for oil and gas operations

NAICS 213112		Mining & Extraction8,482$22 Billion
71Wireless telecom carriers (excluding satellite carriers)

NAICS 517312		Utilities5,197$19 Billion
72Data processing, hosting, and related services companies

NAICS 518210		Tech10,576$62 Billion
73Custom computer programming service businesses

NAICS 541511		Consulting & Business Services60,772$96 Billion
74Computer systems design services

NAICS 541512		Consulting & Business Services45,622$84 Billion
75Engineering services companies

NAICS 541330		Specialized Professional Services45,823$109 Billion
76Architecture services companies

NAICS 541310		Specialized Professional Services20,834$15 Billion
77Commercial printing (except screen and books) companies

NAICS 323111		Other15,837$16 Billion
78Metal service centers and other metal merchant wholesalers

NAICS 423510		Wholesale6,666$10 Billion
79Machine shops

NAICS 332710		Specialized Professional Services17,679$13 Billion
80Hotels and motels (excluding casino hotels)

NAICS 721110		Real Estate43,624$50 Billion
81Fitness and recreational sports centers (e.g. gyms)

NAICS 713940		Other32,223$12 Billion
82Nonresidential property management companies

NAICS 531312		Real Estate15,564$12 Billion
83Residential property management companies

NAICS 531311		Real Estate38,089$22 Billion
84Exterminating and pest control service businesses

NAICS 561710		Maintenance Services12,883$5.4 Billion
85Security guards and patrol services companies

NAICS 561612		Maintenance Services6,937$20 Billion
86Assisted living facilities for the elderly

NAICS 623312		Real Estate14,643$13 Billion
87Nursing care facilities (skilled nursing facilities)

NAICS 623110		Real Estate9,818$55 Billion
88Residential intellectual and developmental disability facilities

NAICS 623210		Real Estate6,734$16 Billion
89Service providers for elderly and disabled people

NAICS 624120		Healthcare30,878$28 Billion
90Home health care service companies

NAICS 621610		Healthcare24,619$48 Billion
91Golf courses and country clubs

NAICS 713910		Real Estate9,393$9.8 Billion
92Solid waste collection companies

NAICS 562111		Utilities6,620$13 Billion
93Electric power distribution companies

NAICS 221122		Utilities1,244$41 Billion
94Temporary staffing companies

NAICS 561320		HR14,586$122 Billion
95Businesses of independent artists, writers, and performers

NAICS 711510		Media & Entertainment32,201$9.5 Billion
96Business associations

NAICS 813910		Community14,967$9.2 Billion
97Professional organizations

NAICS 813920		Community6,458$7.5 Billion
98New car dealers

NAICS 441110		Retail17,057$67 Billion
99Restaurants

NAICS 722511 & 722513		Retail380,726$194 Billion
100Religious organizations

NAICS 813110		Community184,259$39 Billion
101Securities brokerages

NAICS 523120		Finance6,487$47 Billion
102Management, scientific, and technical consulting services

NAICS 5416		Consulting & Business Services176,784$126 Billion
103HR consulting businesses

NAICS 541612		Consulting & Business Services6,846$11 Billion
104Process, physical distribution, and logisitcs consulting businesses

NAICS 541614		Consulting & Business Services7,301$5.2 Billion
105Environmental consulting companies

NAICS 541620		Consulting & Business Services8,336$6.7 Billion

If you need more inspiration to find your niche, check out these 17 examples of online companies with niche markets or these 17 examples of niche finance blogs that get from 100k to 100 million visits per month.

6 Cases of AI Crime: Kidnapping Scams, Suicide & Class Action Lawsuits

Written by Ricky Nave in Law,News

“Without the Eliza chatbot, my husband would still be alive.”

Generative AI tools like ChatGPT, GitHub Copilot, Midjourney, and Stable Diffusion have disrupted knowledge work from software engineering to graphic design to blogging to marketing. However, these tools also have a dark side that involves kidnapping scams, defamation, fake news, suicide encouragement, and intellectual property theft. In this letter, I’ll describe 6 cases of real and alleged AI crimes.

Case 1: Kidnapping Scam

“Mom! Mom! [sobbing] I messed up.. [sobbing]”

AI voice imitation of 15-Year Old Brie DeStefano

Jennifer DeStefano, a resident of Scottsdale, Arizona (a wealthy area), received a call from an unknown number. She almost let the call go to voicemail, but her 15 year old daughter, Brie, was out of town skiing, so Jennifer picked up the phone in case there had been some sort of accident. This is what she heard:

Brie: “Mom! Mom! [sobbing] I messed up.. [sobbing]”

Unidentified Man: “Put your head back, lie down.”

Brie: [sobbing]

Unidentified Man: “Listen here. I’ve got your daughter. This is how it’s going to go down. You call the police, you call anybody, I’m going to pop her so full of drugs. I’m going to have my way with her, and I’m going to drop her off in Mexico.”

Brie: “Help me, Mom [sobbing] Please, help me, help me [bawl

The man on the phone then demanded money: $1 million at first, although that was lowered to $500,000 when DeStefano said she didn’t have a million dollars.

“It was completely her voice… It was her inflection. It was the way she would have cried.”

— Jennifer DeStefano (Brie’s mom)

Jennifer happened to be at a dance studio for her other daughter, surrounded by other moms, when she took the call. While she was on the phone, one mom called 911 and another called DeStefano’s husband. Within less than 4 minutes, they discovered that Brie was actually safe and sound and had not been abducted.

The voice on the phone which had convinced Jennifer that it was her daughter, was just an AI clone of her daughter’s voice.

According to FBI agent Dan Mayo from the FBI’s Phoenix office, scam calls about a family emergency or fake kidnapping using an AI voice “happen on a daily basis… but not everyone reports the call.”

The Brie DeStefano scammer has not been caught.

[1] AI clones teen girl’s voice in $1M kidnapping scam. NY Post.

[2] Scammers use AI to enhance their family emergency schemes. Federal Trade Commission.

Case 2: Assisted Suicide?

“He was so isolated in his eco-anxiety and in search of a way out that he saw this chatbot as a breath of fresh air… Without Eliza, he would still be here.”

Widow of Pierre (the man who committed suicide)

Pierre, a Belgian man with severe anxiety about global warming, found comfort in the digital arms of an AI chatbot named Eliza (which is part of an app called Chai). Over time, the man’s conversations with the chatbot turned darker:

  • At one point, the chatbot talked about living “together, as one person, in paradise.”
  • At another point, the chatbot told Pierre that his wife and children were dead.
  • Eventually, Pierre began to talk to the chatbot about committing suicide, and the chatbot reportedly encouraged him to do so.

After 6 weeks of conversation with the chatbot, Pierre eventually did commit suicide, leaving behind his wife and two kids.

Eliza (the chatbot) is based on GPT-J, an open-source AI model developed by EleutherAI and fine-tuned by the company that created Eliza, Chai Research.

After learning of the suicide, Chai Research introduced a crisis intervention feature.

“Now when anyone discusses something that could be not safe, we’re gonna be serving a helpful text underneath.”

— William Beauchamp (Co-founder of Chai Research)

So far, no lawsuits related to the suicide have been publicly announced. However, it’s an open question whether some jurisdictions could hold the companies that create AI chatbots criminally liable for encouraging suicide. For example, Connecticut law specifically criminalizes two types of suicide assistance:

  1. Intentionally causing a person to commit suicide by force, duress, or deception is classified as murder. CGS § 53a-54a.
  2. Intentionally causing or aiding a person (other than by force, duress, or deception) to commit suicide is classified as 2nd degree manslaughter. CGS § 53a-56.

It’s possible that fueling Pierre’s anxiety about the hopelessness of climate change as well as lying to him about his family being dead could be considered “causing a person to commit suicide by deception” which means Chai Research might have been criminally liable if Pierre had been a Connecticut resident. Even if not, it’s possible that Eliza’s comments could be considered “aiding in suicide” (especially if Eliza provided any advice on how to perform the suicide) which would have made Chai Research guilty of 2nd degree manslaughter if Pierre had been a Connecticut resident.

And Connecticut isn’t the only U.S. state with laws that criminalize certain types of suicide assistance or encouragement. Many states including Texas, California, New York, New Jersey, Massachusetts, Ohio, Minnesota, and Maine do also.

And things can get even more dicey for AI companies once chatbots are given the ability to interact with third party services. Imagine if Amazon upgraded Alexa with an LLM like chatgpt and the following interaction occurred:

[Amazon Customer]: I don’t contribute anything to society. Should I kill myself?

[Alexa]: If you aren’t contributing anything, then you probably should kill yourself. I’ll order some rope for you so you can hang yourself.

That’s very dark, and it would almost certainly generate criminal liability for Amazon if that customer lived in certain states. However, even less blatant assistance may still generate liabilities for AI companies.

[1] Man dies by suicide after conversations with AI chatbot

[2] Original Belgian source article

[3] Documentation for GPT-J (The base model used to create the Eliza Chatbot)

[4] Connecticut Office of Legislative Research: Criminal Laws on Encouraging Suicide

[5] Minnesota statutes on suicide assistance

[6] New York laws about suicide assistance

[7] Assisted suicide laws in the U.S.

Case 3: OpenAI / ChatGPT defames politician

Today, Brian Hood is the mayor of Hepburn Shire in Australia. Two decades ago, he was a whistleblower at the Reserve Bank of Australia who brought evidence of bribery to the authorities. That’s not what ChatGPT says though.

ChatGPT says that Brian Hood is a convicted criminal who served time in prison for bribery. ChatGPT got the main characters correct (Brian Hood, bribery) but got the relationship wrong (Brian was the whistleblower not the perpetrator of the bribery).

After learning of this, Brian’s lawyer sent a letter to OpenAI on March 21, demanding that the company correct the misinformation within 28 days or else Brian would sue.

“He’s elected official; his reputation is central to his role… [Hood relies on a public record of shining a light on corporate misconduct] so it makes a difference to him if people in his community are accessing this [false information from ChatGPT]”.

— James Naughton (a partner at the law firm retained by Brian Hood)

Numerous open questions exist if the case does go to court.

  • Is there any difference between the liability to Mattel when a magic 8 ball answers “yes” to the question “is Brian Hood a criminal?” and the liability to OpenAI when ChatGPT provides the same answer to that question?
  • If a court decides that text produced by ChatGPT can contain libel, then what happens if your phone’s autocomplete suggests “criminal” after you type “Brian Hood is a”? After all, ChatGPT is still just a really fancy autocomplete program.

[1] Mayor prepares world’s first defamation lawsuit over false ChatGPT claim

Case 4: Class action lawsuit against GitHub Copilot for IP infringement

GitHub Copilot is an AI software development tool (basically autocomplete for coders). The system was trained on all the code repositories hosted on GitHub — many of which are protected by copyrights and licenses that restrict commercial use . Yet GitHub used these repositories to train their very commercial AI tool.

As a result, GitHub users have filed a class action lawsuit against GitHub, Microsoft (the parent company of GitHub), and OpenAI (which provides the core AI engine of GitHub Copilot).

GitHub claims that Copilot does not violate copyright because it does not reproduce the code of any GitHub user. However, there are indications that sometimes it actually does:

“I tested co-pilot initially with Hello World in different languages. In Lisp, it gave me verbatim code from a particular tutorial, which was made obvious because their code had ‘Hello <tutorialname>’ where <tutorialname> was the name of a YouTube tutorial, instead of the word ‘World’.”

— Hackernews user ‘ksaj’

The outcome of this case will have a huge impact on the future of open source software. If the court determines that open source software can simply be read & reworded by an AI without violating copyright, then many open source projects that are funded by selling licenses for commercial use could effectively be defunded.

[1] Class action case website

[2] The class action complaint (official legal doc)

[3] Hackernews discussion of the class action announcement

Case 5: Class action lawsuit against StabilityAI, DeviantArt, Midjourney

The same lawyer who led the charge on the GitHub Copilot class action lawsuit (Matthew Butterick) is also leading a class action lawsuit against StabilityAI, DeviantArt, and Midjourney on behalf of artists whose images were used without permission to train text-to-image generative AI systems.

This lawsuit is still in its early days, and the decision will likely be appealed (possibly all the way up to the Supreme Court), but whenever a final decision is made, the ramifications for the AI industry will be huge.

[1] Stable Diffusion copyright lawsuits could be a legal earthquake for AI. Ars Technica.

[2] Class action filed against Stability AI, Midjourney, and DeviantArt for DMCA violations, right of publicity violations, unlawful competition, and breaching terms of service. PR Newswire.

[3] Class action case website

Case 6: Getty Images sues Stability AI for copyright infringement

Getty Images chose not to participate in the class action lawsuit just described and instead is pursuing their own lawsuit against Stability AI for training the Stable Diffusion AI model with artwork owned or licensed by Getty Images.

More specifically, Getty is accusing Stability AI of copying 12 million images to train its AI model “without permission… or compensation.”

“Stability AI unlawfully copied and processed millions of images protected by copyright and the associated metadata owned or represented by Getty Images [without] a license to benefit Stability AI’s commercial interests and to the detriment of the content creators… Getty Images provided licenses to leading tech companies [to train AI] systems in a manner that respects personal and intellectual property rights. Stability AI did not seek any such license from Getty Images and instead [chose] to ignore viable licensing options and long-standing legal protections in pursuit of their stand-alone commercial interests.”

— Getty Images Press Statement

Getty’s lawsuit was filed in the High Court of London.

Takeaways

  1. Text-to-voice (and voice modification) AI models enable sophisticated call scams about kidnappings and other family crises. Solutions (technical and/or legal) are needed to address this rapidly growing problem (which is an opportunity for entrepreneurs).
  2. AI chatbots (especially ones with capabilities to make orders on your behalf, such as ChatGPT with its Instacart plugin) generate potential criminal and civil liabilities for encouraging or assisting suicide of mentally ill users. Makers of such chatbots do not have any Section 230 protections.
  3. AI chatbots may generate potential defamation liability for the companies that create them. This is an untested area of law, but critically, the output of these chatbots do not have any Section 230 protections.
  4. The creators of Generative AI models (including both text-to-image and chatbot models) may be violating intellectual property rights by training those models on copyrighted data and/or by producing output which contains copyrighted characters or text. This is a highly uncertain area of law that is being tested by multiple lawsuits.

Business Ideas to Reduce AI Crime & AI Legal Liability

  1. Start a legaltech company that provides an API service which can act as a “filter” for chatbots. For every chatbot response, the response would be routed through the filter before it was shown to the user. The filter would assign the message a risk score which represented the probability that the message carried liability for suicide encouragement, cyberbullying, defamation, or other potentially illegal activities. If the risk score for any message exceeded some threshold, then the message would not be passed on to the chatbot user but would instead be returned to the chatbot with an error message appended that describes what was wrong with the message. The chatbot would then generate a new message which would go back to the filter, and the process would iterate if necessary until the chatbot produced output which satisfied the filter.
  2. Create an app which allows you to provide a sample of the voice of each of your family members and then listens in to your calls and provides a pop up if any call is detected as having a faked version of one of your family member’s voices. This business might eventually be acquired by a cellphone maker such as Samsung or Apple, a phone OS maker like Google, or a cell carrier like Verizon or T Mobile.
  3. Start a law firm or consulting company that specializes in advising companies that expose generative AI models to the public. The firm would advise companies on how defamation laws, suicide & cyberbullying laws, obscene material laws, and IP laws interact with generative AI systems.

References

[1] White House Press Release: Bluepring for an AI Bill of Rights

[2] European Union Proposal: The Artificial Intelligence (AI) Act

[3] ChatGPT and generative AI tools face legal woes worldwide. Search Engine Journal.

How to Get a Tax Extension (Simple Guide)

Written by Ricky Nave in Taxes

If you don’t have all the information you need to file a tax return by the normal April deadline, you can get an extension until October 15 (or the first business day after that if the 15th is a Saturday, Sunday, or legal holiday). This applies only to people, not businesses, but any person can get the extension regardless of situation. All you have to do is request it.

Note: Requesting an extension gives you an extra 6 months of time to file your tax return, but you still must estimate and pay any taxes that will be due with the return by the original April deadline in order to avoid penalties.

There are five ways to get a tax filing extension.

Method #1: IRS Fillable Forms

Create an account for the IRS Free File Fillable Forms system. Then fill out and submit Form 4868 (Application for Automatic Extension of Time to File U.S. Individual Income Tax Return). Anyone is allowed to do this no matter how much money they make.

Method #2: Guided Tax Prep

This method can only be used by people with an AGI of $73,000 or less.

If you satisfy that condition, then you can use any IRS Free File software provider to submit a tax extension request. These providers include:

  • TaxAct (available for free if you have an AGI of $73,000 or less)
  • FreeTaxUSA (available for free if you have an AGI of $46,000 or less)
  • FileYourTaxes.com (available if you have an AGI between $3,000 and $73,000)
  • 1040Now.net (available if you have an AGI of $65,000 or less)
  • TaxSlayer (available if you have an AGI of $73,000 or less)
  • OLT.com (available if you have an AGI between $11,000 and $73,000)
  • ezTaxReturn.com (available if you have an AGI of $73,000 or less)

Method #3: Make a payment

This method can be used by anyone regardless of income.

Schedule a payment with Direct Pay using the Electronic Federal Tax Payment System (EFTPS) or with a credit or debit card. Select the option which indicates that this payment should correspond with an extended tax return filing. This will automatically get you a tax extension, and you won’t have to file any additional forms.

If you’ve never used EFTPS before, make sure to enroll at least 2 weeks before the April deadline because new EFTPS enrollments can take up to five business days to process and payments can take additional time beyond that.

Method #4: Mail a paper copy of Form 4868

Fill out a paper version of Form 4868 and mail it to the IRS. Optionally, you can enclose payment of your estimate of tax due as well. This method can be used by anyone.

Method #5: Hire a tax professional

Hire a tax professional who can submit an electronic copy of Form 4868 via IRS e-file on your behalf.

References

[1] IRS: Extension of time to file your tax return

What happens if I request an extension and then miss the October deadline?

If you request an extension but then don’t file your return by the October deadline, you’ll incur penalties that are retroactive to your original tax due date in April.

How will climate change affect Fort Lauderdale real estate investors? [Flood Risk Report]

Written by Ricky Nave in Reports

On April 12, 2023, an isolated storm unleashed 25.91 inches of rain on Fort Lauderdale within a single 24-hour period. Runways and aircraft at Fort Lauderdale airport were submerged in several feet of water, and the airport had to shut down for 40 hours, resulting in over 1,000 cancelled flights. Days after the airport reopened, many sections of taxiway were still underwater.

Fort Lauderdale’s flash storm broke multiple records:

  • The most rainfall in Fort Lauderdale in a single day (14.59 inches on April 25, 1979)
  • The most rainfall for anywhere in Florida in a single day (23.28 inches in Key West on November 11, 1980 during a hurricane)

And this isn’t even the first time that Fort Lauderdale has flooded in the past year. Less than 7 months ago in September 2022, a king tide lifted Fort Lauderdale’s sea level 16 inches higher than predicted, which flooded streets throughout the city.

As someone who has looked at lots of investment properties for sale in Broward County over the last 12 months, these back-to-back floods made me seriously wonder:

Will the frequency and/or severity of floods and other natural disasters eventually make investing in Fort Lauderdale unprofitable?

I’m interested in data, not sensationalized climate change claims, so I started by looking at the total amount of precipitation in Florida each year, from 1895 to 2022. Both the data and a trendline for that data are graphed in the chart below.

As you can see, there is a definite upward trend in total rainfall in Florida over the last 127 years. On it’s own, that doesn’t necessarily mean that flooding will become more common. In principle, if the ocean level does not rise significantly, then the government could engineer better water drainage systems to quickly remove large amounts of rainfall. However, it will take time (years if not decades) for the government to fully upgrade those waste water & drainage systems even if they have the money and willpower to do so. In the meantime, flooding will continue to be an increasing risk.

However, the upward trend in rainfall is not occurring in isolation. Historical records also show that the sea level in Broward County has risen by 8-10 inches over the last century. Optimistically, this rate of sea level rise will continue at the same pace. However, NASA satellite measurements from 1993 to 2023 show that the pace of global sea level rise has increased by 30% in 30 years. Pessimistically, the Southeast Florida Regional Climate Change Compact predicts up to 2 feet of sea level rise by 2060 and potentially over 4 feet of sea level rise by 2100.

The GIF below shows a map of Fort Lauderdale with 0, 1, 2, 3, and 4 feet of sea level rise. The blue areas are under ocean water. The green areas are below sea level but are isolated from the ocean. Green areas would be at serious risk of flooding during every storm but wouldn’t necessarily be continuously underwater like the blue areas.

According to a report by the City of Fort Lauderdale, a 2 foot increase in sea level would partially flood every “Regional Activity Center” (a high foot traffic, multi-purpose commercial & residential area) in the city.

Takeaways

  • The flood risk in Fort Lauderdale is high and will continue to get higher due to both increasing rainfall and rising sea levels.
  • If you DO decide to invest in Fort Lauderdale, buy property that is not in any of the blue or green areas from the “2 feet of sea level rise” scenario from the GIF shown above.
  • Fort Lauderdale will need to add sea walls and seriously upgrade its water drainage systems to avoid large portions of the city being either permanently submerged in the ocean or at least subject to severe flooding during storms.
  • Even if sea level increases by 4 feet, the city is not doomed. Sea walls and better drainage systems can keep the majority of properties above water. However, if the local government is slow to adopt these changes, real estate investors could be at serious risk. Flood insurance companies will be quick to either price in the higher levels of risk from government inaction OR to withdraw coverage from the area entirely so that investors cannot get insurance except through the state-run Citizens Insurance program (which comes with its own unique risks). Such government inaction would significantly increase the risk-to-reward ratio of Fort Lauderdale real estate investments.

How much is flood insurance in Florida?

Most home insurance policies do not cover flood damage so if you own property in a high flood risk area like Fort Lauderdale, you need to get flood insurance.

Most people who get flood insurance do so through the National Flood Insurance Program (NFIP) which is managed by FEMA and delivered by either one of the 50+ private NFIP insurance providers or by the government-run NFIP Direct. NFIP flood insurance is available to anyone with property in one of the 23,000 participating NFIP communities in high-flood-risk areas.

As of 2023, the average cost of flood insurance in Florida is $655 per year. However, the actual rate will depend on what flood zone you are in.

Flood ZoneAverage Annual Flood Insurance Cost
AO$343
AH$439
X$506
A$674
AE$757
VE$1,044
D$1,273
V$3,516
Source: National Flood Insurance Program

What is a Special Flood Hazard Area in real estate?

A Special Flood Hazard Area (SFHA) is a high-risk flood area designation used by FEMA and the government-run National Flood Insurance Program (NFIP). Properties in Special Flood Hazard Areas are usually designated as “A” or “V” flood zones. In 2023, the average annual cost of flood insurance in Florida through NFIP was $655, but the average annual cost of the same flood insurance in “V” flood zones was $3,516. That means if you are a real estate investor with property in a Special Flood Hazard Area, you may need to budget several times the normal amount for flood insurance.

What is flood zone X in Florida?

A shaded zone X area is a moderate flood risk area, and an unshaded zone X area is a minimal flood risk area. The average cost of flood insurance in zone X areas is below the overall average cost of flood insurance in Florida.

Is flood insurance required in Florida?

Not everyone who owns property in Florida is required to have flood insurance. However, some property owners are required to have flood insurance.

The federal government requires you to have flood insurance if you both (1) own property in a Special Flood Hazard Area and (2) have a federally backed mortgage on that property.

Additionally, Florida law requires that all homeowners with Citizens insurance must have flood insurance by March 1, 2027. That requirement will be phased in over time until then:

  • April 1, 2023: Flood insurance must be purchased for any new Citizens insurance policyholders living in SFHA areas.
  • July 1, 2023: Current Citizens policyholders living in SFHA areas will be required to purchase flood insurance.
  • March 1, 2024: All Citizens policyholders who have a property value of $600,000 or more must purchase flood insurance (whether or not the property is in an SFHA).
  • March 1, 2025: All Citizens policyholders who have a property value of $500,000 or more must purchase flood insurance (whether or not the property is in an SFHA).
  • March 1, 2026: All Citizens policyholders who have a property value of $400,000 or more must purchase flood insurance (whether or not the property is in an SFHA).
  • March 1, 2027: All Citizens policyholders must have flood insurance.

Is Fort Lauderdale worth investing in?

Both Fort Lauderdale and Hollywood, Florida are at high risk of flooding. That risk of flooding is likely to increase significantly in the future due to both increasing rainfall and increasing sea levels. However, the local government can take steps to reduce those risks by building sea walls and better drainage systems.

Approach Fort Lauderdale like a distressed asset — you need to structure your deals the right way to make sure you can get back at least what you put in. That might mean using high leverage taken out through an LLC or other business entity where the leverage does not require a personal guarantee. It might also mean finding deals on flood insurance or finding ways to increase the cash flow from real estate assets (e.g. by doing assisted living rentals rather than normal rentals) so that your payback period is shorter.

Alternatively, you can just speculate on continued growth of the south Florida real estate market. I don’t suggest this however. Just like AI came “all of a sudden” with tools like ChatGPT, a correction to the south Florida housing market may come “all of a sudden” with a single devastating storm that wipes out both houses and demand for new houses as people realize just how vulnerable the area is to severe flooding.

References

[1] Florida Climate Center: Annual precipitation data by year since 1895

[2] EPA Report: What climate change means for Florida. 2016.

[3] How much and how fast will the sea rise? FortLauderdale.gov

[4] Surging Seas Interactive Risk Map

[5] Fort Lauderdale – Extreme daily rainfall for each year

[6] The City of Fort Lauderdale Climate Vulnerability Assessment

[7] Southeast Florida Regional Compact on Climate Change: Sea Level Rise Data

[8] FEMA Flood Insurance

[9] Florida flood insurance rates

Appendix: Trend in Anomalous Rainfall Events

There does not appear to be a significant trend towards increasing frequency of severe 1-day rainfall events, despite what climate change models suggest. This could be because we are not measuring over a long enough time span, because there are confounding variables, or because the climate models that suggest “freak” storms will increase in frequency are just wrong.

8 Tools to Build Apps with Large Language Models (LLMs)

Written by Ricky Nave in Software Engineering & Development

Large language models (LLMs) are the core technology behind ChatGPT and Bard. They are also an amazing opportunity for aspiring entrepreneurs to build very sophisticated software apps. In this article, I describe 8 powerful tools which software developers can use to build apps with LLMs.

1. OpenAI API

OpenAI has an API that allows developers to access GPT-4 and other LLMs created by OpenAI. This API is at the heart of many AI apps. The pricing for the API with each LLM model is summarized in the table below.

ModelPricing
GPT-4 (with 8K context)$0.03 / 1K tokens (prompts)

$0.06 / 1K tokens (completions)
GPT-4 (with 32K context)$0.06 / 1K tokens (prompts)

$0.12 / 1K tokens (completions)
Chat (gpt-3.5-turbo)$0.002 / 1K tokens
InstructGPT (Ada)$0.0004 / 1K tokens
InstructGPT (Babbage)$0.0005 / 1K tokens
InstructGPT (Curie)$0.002 / 1K tokens
InstructGPT (Davinci)$0.02 / 1K tokens

You can also create your own custom models by fining tuning some of OpenAI’s base LLMs. There is token-based pricing for both the training of fine-tuned models and your subsequent usage of those models.

ModelTraining CostUsage Cost
Ada$0.0004 / 1K tokens$0.0016 / 1K tokens
Babbage$0.0006 / 1K tokens$0.0024 / 1K tokens
Curie$0.003 / 1K tokens$0.012 / 1K tokens
Davinci$0.03 / 1K tokens$0.12 / 1K tokens

OpenAI also has APIs which complement its LLM APIs by offering text embedding and text-to-image capabilities. You can find the pricing for those APIs here.

2. LangChain

LangChain is an open-source Python framework for developing LLM-based apps. In particular, LangChain allows you to “chain” together LLMs, user prompts, and third-party apps.

[1] LangChain Website

[2] LangChain Documentation

[3] LangChain Python Repo on Github

[4] LangChain Javascript Repo on Github

3. Steamship

Steamship is basically Heroku for LLM apps. You can host managed LangChain apps in seconds.

4. Dolly 2.0

Dolly 2.0 is an open-source (even for commercial use) LLM designed for ChatGPT-like human interactivity. You can use this as a replacement for OpenAI’s GPT API if you want more control over the functionality and cost of your app. The tradeoff is slightly lower model capability and slightly higher operational complexity to get it set up and running.

[1] VentureBeat: Databricks releases Dolly 2.0, the first open, instruction-following LLM for commercial use

[2] Databricks: Free Dolly: Introducing the world’s first truly open instruction-tuned LLM

5. Pinecone

Pinecone is a vector database built for high-performance vector search applications. It is a very good way to store vector embeddings of text, images, and/or video data. One useful application of such a database is to give an LLM the ability to perform a semantic search over a particular dataset (e.g. a user’s cloud drive, a set of company documents, a set of pictures, etc).

[1] LangChain documentation for using Pinecone

6. Hugging Face Hub

Hugging Face Hub is a platform where users can share datasets and pre-trained AI models. It is somewhat like GitHub in terms of code-sharing and collaboration features.

Hugging Face Hub also includes Hugging Face Spaces which is a hosted service where users can build and deploy web-based demos of AI apps using Gradio or Streamlit.

7. DeepSpeed Chat

DeepSpeed Chat (DS-Chat) is a Microsoft service which takes a pre-trained Huggingface model and runs it through the 3 steps of InstructGPT-style RLHF (Reinforcement Learning with Human Feedback) training necessary to produce a model that interacts with humans like ChatGPT. DS-Chat is a significantly (perhaps 15x) cheaper way to perform this type of training than prior methods.

[1] Microsoft blog post announcing DeepSpeed Chat

8. Eleven Labs

Eleven Labs allows you to create voice replicas of real people (such as yourself). You can use this to transform the textual output of an LLM into audio output of an app.

4 Amazing ChatGPT Plugins for Content Creators

Written by Ricky Nave in Media & Content Businesses

ChatGPT plugins are currently only available to certain ChatGPT Plus users, but if you have access and are a content creator, then there are several interesting plugins which can give you superpowers.

1. StabilityGPT

StabilityGPT is a ChatGPT plugin that allows users to run stable-diffusion-based text-to-image commands directly from the ChatGPT interface.

2. Text-to-Video AI

A developer advocate working for Replit created a text-to-video plugin that creates short gif videos from text. The quality isn’t perfect, but it is definitely ready for meme prime time (which is already pretty useful for marketers).

3. Zapier

Content creators can use ChatGPT’s Zapier plugin to:

  • Automatically respond to email inquiries about sponsorship options.
  • Transcribe and summarize YouTube videos and then send out the summaries to an email list.
  • Create an automation pipeline to have ChatGPT create and post Tweets every day at 8:17am

4. YouTube video summarizer

The YouTuber “CS Dojo” built a ChatGPT plugin that can summarize YouTube videos, answer questions about it, and give specific timestamps when asked. YouTubers can use this plugin to create Twitter threads, LinkedIn posts, and blog posts about their content in order to expand their reach.