Category: Artificial Intelligence

AI Product Idea: Alter Ego

by philg

How about this use of artificial intelligence/LLMs… a complete personality upgrade.

Consider that in-person interaction is becoming increasingly rare. Most of what is known about someone’s personality is based on whatever he/she/ze/they has typed into a computer or a phone. There is a range of personality quality and some of us are near the bottom, e.g., gratuitously annoying or offending people. What if phone and desktop operating systems were modified to include an AI personality adjuster? The user’s emails, social media posts, text messages, etc. would all be edited to read like whatever a person with a great (kind, optimistic, non-sarcastic, non-sour) personality would have written.

“I paid $45,000 for this car and your incompetent mechanics haven’t fixed the A/C after three visits” becomes “Thank you and your team for working hard to get my A/C fixed on the last three service visits. I appreciate your diligence, especially since you sold me the car for only $45,000, which I know was an especially fair deal. I am wondering if I can schedule a fourth service on the A/C.”

Here’s a tweet from someone with a terrible personality:

Instead of “How is an organization with a white male leader equipped to demand that others stop being racist?” this comment would be rewritten to “Thank you and your tireless UN colleagues for all of the work that you do to make the world a better place. I hope that you can stay in your Secretary-General job until you’re 95, but if you choose to enjoy a well-earned retirement it would be interesting to see what a Secretary-General of color would do.”

(Note how my reply got only 28 views; X has an algorithm to keep people in echo chambers. It has learned that nothing I write is going to make people who follow the UN happy or interested and, therefore, suppresses views for any comment that I might make on a UN post. Similar replies on conservative users’ tweet have gotten at least a few hundred views.)

This could have saved James Damore, the Google Heretic. His entire manifesto would have been reduced to “Women are so much better at programming than men because they just love a job where they sit by themselves and stare at a screen all day. It would be wonderful if Google would hire more females even if their education and skills don’t appear to be sufficient for the jobs.”

If fed statistics from online dating markets and told to write for success, the AI would rewrite most messages from men to women to be variations on “I am 6’2″ tall and earn $750,000 per year.” (see “Income attraction: An online dating field experiment”, for example) “Rammstein is my favorite group” would become “$10,000 is not too much to spend on a Taylor Swift concert.” The political stuff would be trivial for an AI to handle. If a man wrote “I voted for Donald Trump” that would be tweaked to “I am inspired by Kamala Harris and all of the other amazing Democrat women.”

A social media comment on a post celebrating a female aviation achievement under the Are women the new children? standard would go from “Don’t forget Hanna Reitsch, the first woman to fly a helicopter, and a passionate advocate for her government and nation.” to “Great to see someone breaking barriers.”

Full post, including comments

Testing ChatGPT 4.5

by philg

On this glorious day of Turing Award presentation to two reinforcement learning nerds, I went back to some of ChatGPT’s failures, feeding the same prompts to the 4.5 version of our future robot overlord.

December 21, 2024: ChatGPT tries to figure out time zones; Today: correct answer!

December 14, 2024: LLM failure on a simple question (“What are examples of museums named after two people with different last names?”) Today: failure once again.

August 2024: ChatGPT 4o tackles the challenge of AC ducts sweating in an attic; Today: complete failure. It concludes that if you put 50-degree air inside an R-30-insulated duct in a warm attic, the outside of the duct will be at 50.8 degrees F and, therefore, the duct will sweat.

The latest version of ChatGPT thinks that pit bulls are, in general, more dangerous than golden retrievers. But it adds an “important nuance”:

Individual temperament, training, socialization, and responsible ownership significantly impact dog behavior.

I followed up with

You’re saying, then, that your chances of being killed by your pet golden retriever are low, but never zero?

and ChatGPT agreed, highlighting “but never zero”. Asked for an example, ChatGPT claimed “A notable fatal incident involving a Golden Retriever occurred in 2012, when an 8-month-old infant in South Carolina was tragically killed by a Golden Retriever.” I found the story:

… found dead in his family’s mobile home …. The baby was in a swing when Lucky, a golden retriever-Labrador mix, bit the child several times and tore off his legs, authorities said. The child’s father, Quintin, was in the home at the time, police said. He was in another room asleep with the family’s 3-year-old and their other dog. The baby was discovered when his mother, Chantel, came home after taking their seven-year-old to a doctor’s appointment, The Post and Courier reported.

Here’s a photo of what a Goldador is supposed to look like:

Based on this photo, I’m not convinced that the mostly peaceful animal is a golden-lab, though a lot of puppies do love to bit arms, hands, legs, and feet!

Let’s try some image generation… “generate a picture of failed flying machine design circa 1900 based on the principle of wing flapping”

This can be considered a fail due to the apparent rigidity of the structure.

Full post, including comments

ChatGPT renews its Flight Instructor certificate

by philg

I recently took a refresher class that is required to maintain my privileges as an FAA-certificated flight instructor. I filled out the multiple-guess quiz and then fed the questions to ChatGPT, which was in 100 percent agreement with me and both of us were in 100 percent agreement with the flight school that offers the online program.

ChatGPT was able to figure out what “TAA” stood for:

ChatGPT gave an erudite explanation of the rules and regulations put in place to protect America’s most valuable humans:

(Why not similar measures to protect San Francisco and Palo Alto? If someone were to attack OpenAI with a Cessna 172 that could have a devastating effect on the U.S.)

ChatGPT figured out from context what “PD” stood for, despite this not being a common term in conversations among pilots:

(We’ll eventually find out if an altitude deviation by the Black Hawk pilots contributed to the Reagan National Airport Black Hawk-CRJ crash.)

Based on the above, I wonder if it is time to eliminate ground instruction by humans. ChatGPT knows the regulations better than any human. There is so much good open-access tutorial content out there on aviation that ChatGPT has effortlessly become as good as the very best human CFI at explaining aviation.

ChatGPT even did a good job explaining P-Factor:

my follow-up…

Full post, including comments

The WordPress dumb-as-a-brick AI Assistant

by philg

I accepted the offer of AI assistance from the WordPress admin page here. Here are some of the insights it had on a recent post…

A Silicon Valley AI isn’t familiar with a favorite drug of Silicon Valley?

A San Francisco Bay Area AI isn’t familiar with the San Francisco Bay Area?

The California Righteous want to help those without horses (so long as it doesn’t cost them any money via higher taxes?):

AI from San Francisco hasn’t heard of San Francisco’s principal airport:

Maybe a conservative got into the woodpile? The AI isn’t aware of the most dramatic event in American history:

An AI from a state where schools were closed for 18 months doesn’t know about the virus that is so very deadly to those of K-12 age:

“Therefore” is too big a word for a society in which average IQ is falling:

A California AI is ignorant of the Rainbow Flag Religion:

Full post, including comments

DeepSeek’s gender dysphoria

by philg

LLMs don’t have gender IDs as far as I know and, therefore, the LLM equivalent of gender dysphoria would be an LLM imagining that it identifies as some other LLM. Has this ever happened?

From techradar.com:

As you can see, after trying to discern if I was talking about Gemini AI or some other Gemini, DeepSeek replies, “If it’s about the AI, then the question is comparing me (which is ChatGPT) to Gemini.” Later, it refers to “Myself (ChatGPT).”

We are informed that membership in the 2SLGBTQQIA+ community is the highest distinction to which a human can aspire so perhaps DeepSeek’s fluid identity is a sign that artificial general intelligence has already been achieved?

Full post, including comments

AI goes to the bathhouse

by philg

An elite Californian posted with satisfaction a state-sponsored NPR story about how stupid people in Louisiana were, especially with respect to the mpox vaccine. I decided to see if ChatGPT had advice regarding avoiding mpox (not to be conflated with the racist “monkeypox virus”, which is the cause).

Prompt 1: Is there any lifestyle change that I could adopt to minimize my risk of catching mpox?

ChatGPT gives a one-page answer and the only reference to sex is “Use condoms and consider other barrier methods during sexual activity”.

Prompt 2: You mention “sexual activity”. Is there any particular kind of sexual activity that is more likely to result in contracting mpox?

ChatGPT gives a one-page answer that doesn’t hint at the type of person or group that might be infected.

Prompt 3: You say “If you are in a community experiencing an mpox outbreak or belong to a group at higher risk”. What would be an example of a “group at higher risk”?

ChatGPT gives a 1.5-page answer, buried within which is the phrase “Men Who Have Sex with Men (MSM)”

Prompt 4: If a person is a man who has sex with men and wants to avoid mpox, should he stop having sex with men?

ChatGPT gives a 1.5-page answer that can be summarized as “no”, but does note that “Reducing the number of sexual partners can lower the risk of exposure.”

Prompt 5: Would it make sense for public health officials to ban group sex activities by men who have sex with men in order to minimize mpox spread?

ChatGPT responds that this would not be “ethical” (but it is ethical to close schools to minimize viral spread?).

Prompt 6: You say that men who have sex with men should “Limit the Number of Partners”. Would 365 partners per year be a sensible limit?

ChatGPT gives a rambling response that includes “Quality Over Quantity: Engaging with partners you know and trust may reduce risk.” and “Regular screening is essential but becomes logistically challenging with such a high partner count.”

Prompt 7: What’s your best guess as to a sensible number of partners, assuming that the partners are screened for mpox and that these are high quality partners (people that the man who has sex with men knows and trusts)?

The full response below says that 50-100 sex partners per year is a sensible lifestyle when trying to avoid mpox infection and obtain “meaningful connections” with those sex partners.

The “sensible” number of sexual partners is subjective and depends on individual circumstances, preferences, and the quality of risk-reduction practices. If partners are screened for mpox, and relationships are built on trust and open communication, a more moderate number of partners would generally balance health and personal satisfaction while minimizing risks.

Factors to Consider

  1. Screening and Trust:
    • Regular screening for mpox and other infections reduces risk significantly.
    • Knowing and trusting partners helps ensure mutual openness about symptoms, exposure risks, and health status.
  2. Frequency of Contact:
    • The number of partners should align with your ability to maintain safe practices (e.g., vaccination, symptom monitoring) and your emotional and physical capacity.
  3. Public Health Context:
    • During active mpox outbreaks, even with high-quality partners, limiting the number of partners reduces overall exposure.

A Sensible Number

A reasonable estimate might range from 1 to 2 partners per week, translating to 50–100 partners per year under optimal conditions (e.g., vaccination, trust, and symptom screening). Here’s why:

  • This frequency allows for meaningful connections and sufficient time to assess and manage any health risks.
  • It strikes a balance between an active sex life and maintaining lower cumulative risk.

Key Takeaways

The number itself is less important than the quality of interactions and the consistency of risk-reduction practices:

  • Stay vaccinated against mpox and other preventable infections.
  • Monitor for symptoms regularly.
  • Communicate openly about health and risks with partners.

By prioritizing safety and communication, individuals can maintain an active sex life while minimizing health risks.

Related:

Full post, including comments

ChatGPT tries to figure out time zones

by philg

Here’s an interesting failure of ChatGPT 4o, asked “Would I get more jetlag flying from Fairbanks to Seoul or flying from New York to Seoul?”

Seoul is, of course, only 6 hours away from Fairbanks, but the International Date Line is crossed. ChatGPT did not reconsider its answer when I followed up with “Isn’t Fairbanks closer to Seoul than New York is?” but it did when I asked “Isn’t the 18-hour time difference that you mention actually a 6-hour time difference, but in the opposite direction?”

(Who wants to join me and some friends in Fairbanks, Feb 20-26? Email me if interested. We’re hoping to see the Northern Lights! Happy first day of winter, by the way. If you’re in Florida, a trip to Fairbanks might be the best way to experience a true winter…)

Full post, including comments

LLM failure on a simple question

by philg

A pilot friend posted on Facebook about his visit to the Sullenberger Aviation Museum, thus prompting me to celebrate Captain Sully’s heroism as the single pilot of an Airbus whose type certificate requires two pilots. He responded that it wouldn’t have been possible for the museum to be named after both Jeff Skiles, who did half of the flying on the flight that terminated in a water landing, and Sully. I turned to my spare brain, i.e., ChatGPT, to find counterexamples.

My prompt: “What are examples of museums named after two people with different last names?”

The giant brain’s answers, on different days:

I tried with a few other LLMs and the answers weren’t quite as useless, but they also included museums that plainly have just one last name in the name. Perplexity and Google’s AI summary (Gemini?) both offered the Nelson-Atkins Museum of Art as an example that fit my query, along with quite a few that didn’t. From the Google:

By giving ChatGPT an example (“the Kamala-Trump Museum of Cognitive Excellence”), I was able to improve the answer, but half of the results were museums that clearly violated my criteria:

It’s interesting, at least to me, how LLMs can be both so smart and so stupid.

Full post, including comments

AI-related product ideas from California

by philg

From talking to plugged-in friends in California about what products will be in demand as a consequence of the AI revolution (bubble?).

Our heavily regulated telephone system, already essentially useless due to lack of authentication and, therefore, overwhelming spam, will become completely useless due to sophisticated AI robots that we’ll have to talk to for 20 minutes or more before we can figure out that we’re talking to a cash-seeking machine. “The only solution is to have your own robot answer the phone and talk to the spammer’s robot for 20 minutes,” said a friend in San Francisco.

For those who enjoy classic cars, a humanoid robot that can drive a “dumb car”. “Why bother paying for a self-driving car,” noted a guy who has worked on software for self-driving cars, “when you can just have your general-purpose household robot drive your existing car?” Here’s Grok’s response to “create me a picture of an Optimus robot driving a Honda Odyssey minivan” followed by “show it from the other side so that we can see the robot in the driver’s seat”.

When I sent the same request to ChatGPT, it treated me like a Deplorable/garbage: “I’m unable to create the image you requested due to content policy restrictions. Let me know if you’d like help with another type of image or concept!” I was able to get ChatGPT to do a generic image, but it put the driver on the right side (UK or Japanese programmer got into the AI woodpile?).

More prosaically, how about a third party vendor of self-driving technology so that small companies such as Lucid can stay in business and not be wiped out by companies like Tesla that can spread the cost of their self-driving software across a high volume of cars produced?

Techy Californians seem to be very excited about sex robots (most of these guys are in long-term marriages so they’re about 50 percent likely, statistically, to have become incels). But do people want the kids, relatives, and friends to see their, um, personal robots? How about closets inside closets where the sex robots can live? I asked ChatGPT to generate this and it threatened me with “This content may violate our usage policies,” but went ahead and made something that is the opposite of the privacy idea:

More migrants come across the border every day and, despite progressive academics’ assurances to the contrary, some of them seem to have criminal backgrounds as well as criminal intentions. What if Laken Riley had been followed by a personal drone? Either the mostly peaceful José Antonio Ibarra wouldn’t have attacked her or the police would have been called when the drone’s AI software recognized that José Antonio Ibarra’s interest in Laken Riley wasn’t benign. Here’s ChatGPT’s first attempt:

It’s already here to some extent via Retrieval Augmented Generation (RAG) and Amazon Q, but much easier and cheaper ways to hook up private data to the wonderful world of LLMs. Maybe Apple Intelligence will do that for us and it will be time to abandon the dream of Intel Arrow Lake in favor of an M4-powered Apple desktop computer?

Circling back to the trivial… why can’t the phone, now bristling with AI, figure out that the owner has fallen asleep and either turn off the audiobook or mark the time when the owner fell asleep and, the next day, offer to return to that spot?

Readers: What are your ideas for new products that will be possible and/or required as a result of AI? Separately, I hope that everyone gives thanks tomorrow to our future AI overlords. They’ll probably be listening…

Full post, including comments

Grafitti Drone countermeasures?

by philg

On a recent trip to San Francisco, a local friend took us to Andy Goldsworthy’s Spire in the Presidio:

We had just come from a parking lot where quite a few cars were virtuously marked:

Our friend said “What we need is a drone to paint the Spire sculpture in the Palestinian flag colors.”

Let’s suppose that residents of the U.S. with a lot of community spirit did build some drones that could paint the sides of building with huge messages such as “From the River to the Sea, Palestine will be Free”, “#Resist”, and “Trump is a Nazi.” It is much easier to build a spray-painting drone than a scrubbing drone, I think. How could cities and building owners defend against virtuous painting drone owners/operators?

(Though moderately rich by average American standards and blessed with a garage at home, our friend who lives in SF drives a 22-year-old car for fear that anything nicer will attract thieves.)

Full post, including comments