OpenAI was allowed to relaunch ChatGPT in Italy after making changes demanded by Italian data privacy regulators.
ChatGPT was originally banned in late March after the Italian Data Protection Authority complained that ChatGPT did not comply with GDPR and that ChatGPT provided inappropriate information to minors.
OpenAI has now implemented most of the changes demanded by Italian regulators:
OpenAI created a form that allows EU citizens to request their personal data be removed from ChatGPT responses.
OpenAI published a disclosure of the personal data that is processed for ChatGPT’s training, and a reminder that everyone has a right to opt out.
OpenAI added an age gate in Italy that asks users to confirm that they are either at least 18 years old OR are at least 13 years old and have obtained consent from a parent or guardian, before that person can use ChatGPT.
In response, Italian regulators issued a temporary approval for ChatGPT to be again be offered in Italy again, with the condition that OpenAI continue to work on implementing additional data protections into ChatGPT over time.
“The Italian SA acknowledges the steps forward made by OpenAI to reconcile technological advancements with respect for the rights of individuals and it hopes that the company will continue in its efforts to comply with European data protection legislation.”
Announcement by Italy’s Guarantor for the Protection of Personal Data
This is still just the beginning of OpenAI’s regulatory conflicts though as governments around the world are grappling with how to balance technological advancement with the legal quagmire of AI.
The entire forestry & logging industry in the U.S. has 6% of Google’s revenue and about 50% of Salesforce’s revenue. That’s how important it is to choose the right market for a new business. Big markets give a new business unbounded room to grow, and small niches within big markets have all the advantages of small markets but without the growth limitation.
This means, for example, that it would probably be a much better idea to create a company that sells life insurance to loggers than to create a logging company. The former could expand into other types of life insurance and/or other types of insurance — a HUGE industry. The latter would be capped in potential even if you took over the entire logging industry. You could of course always expand from logging into construction, but the two are not as similar as life insurance for loggers and life insurance for non-loggers. Plus, even the construction industry isn’t as big as the insurance industry.
Total revenue is one way to measure the size of an industry, but GDP is arguably a better way. The table below summarizes the GDP of 17 industry segments in the U.S. economy in 2022.
Industry
Examples of Included Business Types
Number of Distinct Companies with at Least 1 Employee
– Newspapers – Book publishers – Software publishers – Greeting card publishers – Movie production companies – Movie theaters – Music publishers – Recording studios – TV broadcasting – Wireless phone service carriers – Cloud computing – Data hosting
82,998
$1.39 Trillion
10
Construction
NAICS 23
– Residential building construction – Industrial building construction – Utility system construction – Oil and gas pipeline construction – Highway and bridge construction – Framing contractors – Masonry contractors – Glass contractors – Roofing contractors – Siding contractors – Plumbers – HVAC companies – Drywall contractors – Painters – Concrete contractors
Create a Chrome browser plugin that will summarize the key details of a privacy policy or “terms of service” agreement in a single paragraph. The plugin could be monetized by charging a small one-time fee for it (e.g. $2) or it could be used as a customer acquisition tool for other consumer legal-tech tools.
2. AI Newsletter
Create an email newsletter that covers some aspect of AI developments. There is already something of a gold-rush to create AI newsletters, with some of the most successful ones shown in the table below.
In other words, there are more than enough AI newsletters for tech workers.
You know what there isn’t enough of?
Niche, industry-specific newsletters which aren’t AI-only newsletters but rather newsletters which discuss AI developments that are relevant to a people in a particular industry or job role. For example:
Entrepreneurial dentists (There are over 124,000 dentist businesses in the U.S.)
Entrepreneurial doctors (over 157,000 doctor’s office companies in the U.S.)
Entrepreneurial veterinarians (roughly 28,000 veterinarian businesses in the U.S.)
Lawyers (There are over 161,000 law firms in the U.S. with a collective payroll of $110 billion)
Insurance Agents (over 122,000 insurance agencies and brokerages in the U.S.)
Real estate agents (roughly 121,000 real estate agencies and brokerages)
Management consultants (over 80,000 companies in this niche)
I just talked about how to create an email newsletter about AI for various niches. You can pursue essentially the same strategy to create a YouTube channel instead.
4. Property management for independent landlords
A significant proportion of a property manager‘s workload is messaging — messaging tenants, messaging contractors, messaging handymen. The majority of that messaging can be automated with a tool like GPT-4.
There is an opportunity to create a tool that allows property managers and/or independent landlords to do exactly that. And this opportunity is sizable. There are over 38,000 residential property management companies that pay out over $22 billion in annual payroll.
5. Airbnb messaging
As just mentioned, messaging is a significant part of the workload of a residential property manager. For an Airbnb host, there is even more messaging work involved. There is an opportunity to either create a software tool that automates the majority of Airbnb messaging for hosts OR to offer full-service Airbnb management and use AI messaging internally to decrease your costs (and increase your margins).
6. WWJD-GPT
What would Jesus do? Ask WWJD-GPT!
7. Automated phone system
Many solopreneurs and small business owners do not have the bandwidth to man their phone 24/7 (or even 8/7). Landscapers are a great example of this. Most landscaping companies are 1-man businesses, and those men can’t open their phones when they are mowing grass or trimming hedges. That means there is an opportunity to pipe together a tool like GPT-4 with voice-to-text and text-to-voice models in order to create an automated phone system that can take customer and potential customer calls.
8. Insurance claims adjustment & investigation
There are over 285,000 people who work as insurance claims adjusters, examiners, and investigators in the U.S. The average annual wage is over $73,000 which means insurance companies pay out roughly $21 billion each year for this work. However, a significant amount of this work is done remotely and is amenable to automation using large language models and image analysis AI tools.
9. Software development agency arbitrage
There are over 60,000 businesses under the classification of “custom computer programming service businesses” (NAICS 54151) in the U.S. These businesses pay out over $96 billion in payroll annually, so there is a large opportunity here.
10. Automated Email Customer Support
Many businesses have an email account for customer support. Build an AI tool which can automate this (at least partly — even if 20% of customer support cases have to get elevated to a human, that’s an 80% reduction in customer support workload).
11. AI Textbot Therapist
Have GPT-4 act as a therapist and provide it as a phone number that people can text an unlimited number of times for a monthly subscription.
12. YouTube Thumbnail Generator
Build an AI tool that can analyze a video and automatically produce an amazing, high-performing thumbnail.
13. End-to-End AI-Generated YouTube Channel
Create a YouTube channel that is entirely automated:
Recurring and automatic generation of video ideas based on scraping news and other articles and related automated research processes conducted by GPT-4.
Generate a video script using GPT-4.
Produce an audio track from the script using a tool like Eleven Labs.
Automated selection and sequencing of stock video clips relevant to different segments of the audio track. Once AI text-to-video tools progress far enough, stock video clip selection can be fully or partially replaced by AI video generation.
Automatic Thumbnail generation.
Automatic video description generation.
Automatic upload of video, thumbnail, and description to YouTube.
When selecting a niche for this YouTube channel, you don’t want niches which rely heavily on visuals. That means no tutorials and no Charisma-on-Command-style video analyses. On the other hand, Top-N lists and explanations of concept topics like economics and history could work well.
Channel Ideas
Human geography (how different geographic regions affect and are affected by humans). For example, videos on how the Nile delta affected early human development, how technology developed on remote Pacific islands, or how the availability of various mineral ores affected the rate of industrial development in different countries.
Investopedia. Make a channel that has lots of short videos explaining various finance topics. Investopedia’s YouTube channel currently does not have as much topic coverage as the Investopedia website.
Finance & economics daily news.
Astronomy & astrophysics.
Space exploration & tech.
Ocean science & technology.
Oceans (maritime law, trade routes, listicles of biggest/best ships, ocean history stories, ocean aquaculture, ocean mining, naval warfare & geopolitics).
Collecting money (the history of money, coin collecting, ancient bill collecting, different monetary systems, etc).
Sports history & facts (this can be niched down for American football, baseball, soccer, tennis, golf, pickleball, swimming, track and field, or any other sport).
Bladed weapons (e.g. history, science, smithing, and sports related to swords, knives, and other bladed weapons).
Scary stories. This can be a general scary story channel or can be niched down (e.g. stories about side effects of medicines, legends about lake monsters, stories about innocent people being sent to prison, etc).
Bible verses & stories.
History of clothing.
Philosophy.
Gossip about politicians.
Weird laws.
Human futurism (news & concept explanations about cybernetics, genetic engineering, epigenetics, life extension, etc).
Geopolitics (concepts & news).
Neuroscience.
Etymology.
The history of brands. Tell the stories of how brands were built.
It may be best to target long-tail YouTube search traffic where there is little competition so that viewers will tolerate a lackluster synthetic voice.
14. SaaS product to help people launch AI YouTube channels
Instead of launching an end-to-end AI generated YouTube channel yourself, you could create a SaaS product that allows others to create such channels.
The person specifies the niche and follows a simple user journey to design a YouTube brand at a high level (colors, voice accent, etc). They may also specify the list of sources to check for new articles to use as content inspiration. The software then sets up an automation pipeline that will scrape the sources on a periodic basis for new content, decide which articles to use, scrape them, process them through a large language model into a video script, match up the video script with images and video clips (stock clips and AI generated clips), and then output an entire video. The software will then generate a thumbnail and will post the video to YouTube with an LLM-generated description.
April 2023 is a rough time to be in the media industry.
April 20: Insider, Inc. (formerly known as Business Insider) announced it would lay off 10% of its employees.
April 20: According to an internal memo, Buzzfeed News plans to shut down after the online publication failed to meet profit expectations.
April 23: Executive Jeff Shell agreed to leave NBCUniversal after a female colleague filed a complaint of inappropriate conduct against him.
“I had an inappropriate relationship with a woman in the company, which I deeply regret. I’m truly sorry I let my Comcast and NBCUniversal colleagues down.”
Jeff Shell
April 24: Tucker Carlson was fired from Fox News. This comes less than a week after Dominion Voting Systems won a $787.5 million settlement from Fox for defamation.
April 24: Don Lemon was fired from CNN after 17 years with the network. This comes two months after the news anchor drew criticism for saying that 51-year old Republican presidential candidate Nikki Haley wasn’t in her “prime”.
April 24: ESPN+ streaming executive Russell Wolff was laid off along with 7,000 other Disney workers.
44 Instrumental Background Songs for Instagram Reels
If you want to make informational Reels for Instagram or Facebook, you may want to add undistracting instrumental background music. In this article, I list 44 songs that you can use for this purpose, all of which are available directly within Instagram.
Song Name
Song Mood
Artist
1
Awkward & Funny Instrumental Music
Funny
Jony Pham
2
Hvitserk’s choice
Dark / Epic
Trevor Morris
3
Earth Song (Orchestral Version)
Sad / Beautiful
Akshay Sreeram
4
Mischief (Instrumental)
Unintrusive rap beat
Zen-Beatz
5
Aggressive Wilderness
Fight background music
Peter Jeremias
6
Documentary
Intellectual
Coma-Media
7
Where Is My Mind
Sad / Lonely
Maxence Cyrin
8
Classical Violin
Intellectual / Fast
Rafael Krux
9
Osaka
Asian / Rap
Copyright Free Beats, hozoneonfire
10
Lucky’s Song (Happy Royalty Free Music)
Happy / Child
ContiMusic
11
Vacation Vibe (Popular Royalty Free Music)
Vacation / Tropical
ContiMusic
12
Secret Love
Cool / Uniform / Rap
Copyright Free Beats, hozoneonfire
13
Essence (Instrumental)
Unintrusive
KPH
14
I Like the Flowers – Public Domain Song
Unintrusive / Fun
Ronny Matthes
15
Indian Ethnic Classical
Indian
Sandeep Khurana
16
Fairy Fountain
Relaxing / Guitar
Super Guitar Bros
17
Such a Whore (Instrumental)
Club / Slow but deep
JVLA
18
Mystery Background
Cinematic Mystery
Petar Milinkovic
19
Maza (Instrumental)
Arabian / Upbeat
Ultra Beats
20
Hip-Hop (Instrumental)
Fast forward on autopilot
dead prez
21
No Time for Caution
Cinematic / Epic
Hans Zimmer
22
Fulfilling Humming
Calm / Peaceful
Nasheed
23
Upbeat Background Music
Upbeat / Generic
HitsLab
24
Epic Emotional Dramatic Voices
Epic
Carlos Estella
25
#20
Calm / Sad
Aphex Twin
26
Riptide (Instrumental Version)
Pop Guitar
Soft Background Music, Guitar Tribute Player
27
Food Background Music
Fast / Upbeat / Advertising
HitsLab
28
Special Forces
Action
Shashwat Sachdev
29
Elevator Background
Jazz / Elevator Music
HitsLab
30
Rich Flex (Instrumental)
Steady beat / YouTube
Fruity Covers
31
First Snow
Generic
Giulio Cercato
32
The Final Countdown (Remix)
Hype
Evoxx & Gustavo Koch
33
Trumpet Tune: Trumpet Tune
Royal
Westminster Abbey Choir, London Brass, Martin Neary, Martin Baker
These are the 10 companies which generated the most operating profit out of all companies in the world in 2022. Bank profits have been adjusted to account for unrealized losses which is why no banks made the top 10 list this year. On the other hand, the war in Ukraine skyrocketed the prices of oil and natural gas which doubled or tripled the operating profit of some energy companies.
Saudi Aramco is an Arabian oil and gas company that consistently generates the largest annual profits of any company in the world. The company generated $305 billion of operating profit in 2022 (which is more than the total profits of Apple, Microsoft, Google, and Visa combined). In a single year, the company paid out $75 billion of dividends which is higher than the market cap of Airbnb.
Roughly 95% of the shares in the company are owned by the Saudi government, and Saudi Aramco in turn owns a 70% majority stake in Saudi Basic Industries Corporation (SABIC).
2. Apple ($119 Billion)
Apple generated $119 billion of operating profit. Twenty years ago in 2002, the company generated less than $70 million in profit.
3. Microsoft ($83 Billion)
Microsoft generated $83 billion in operating profit with the majority of that coming from cloud services and MS Office.
4. Alphabet ($79 Billion)
Alphabet generated $79 billion in operating profit with 99% of that coming from Google.
5. Equinor ($79 Billion)
Equinor is a Norwegian government-controlled oil & gas company that generated $79 billion US dollars of operating profit in 2022.
6. Exxon Mobil ($75 Billion)
Exxon Mobil’s operating profits grew 150% from $30 billion in 2021 to $75 billion in 2022.
7. Shell ($67 Billion)
Shell rode the wave of higher oil prices in 2022 to generate $67 billion of operating profit.
8. Petrobras ($57 Billion)
Petrobas is a Brazilian state-owned oil company that generated $57 billion of operating profit in 2022.
9. Chevron
Chevron generated $49 billion of operating profit in 2022, with the vast majority of that coming from oil and gas extraction rather than refinement.
10. TotalEnergies
TotalEnergies is a French multinational oil company that generated $46 billion of operating profit.