Essay Grading - Human vs. Machine

For my daughter’s high school, I volunteered to read and score scholarship applications that were submitted by graduating seniors. There were 10 categories of applications including Academic Excellence, Arts, Athletics, Leadership, School Service, and others. All applications consisted of an essay, and some of the categories required the submission of supplemental information such as photos, videos, or other information to support the applicant’s scholarship candidacy. Parent volunteers were placed in groups of 3, with each group asked to review 4 or 5 applications. Parents were provided with a grading rubric and were asked to independently evaluate each student’s submission. To reduce the chance of bias, parents were asked to be reassigned to another group if they knew the student.

The grading rubric consisted of 5 dimensions for a total of 20 points:

Followed Directions

2 points – Followed most or all directions

1 point – Followed some directions

0 points – Followed no directions

Answered Essay Prompt

3 points – Answered the prompt completely

2 points – Mostly answered the prompt

1 point – Somewhat answered the prompt

0 points – Essay has nothing to do with the prompt

Well-Written and Use of Good Grammar

5 points – Essay is well-written and almost all of the grammar is correct

4 points – Essay is somewhat well-written and most of the grammar is correct

3 points – Essay is adequately written and the grammar is somewhat correct

2 points – Essay is sloppily written and has numerous grammatical errors

1 point – Essay is poorly written and has many grammatical errors

0 points – Essay is incomprehensible

Provided Examples of Supporting Evidence

5 points – Completely supported essay with examples of evidence

4 points – Mostly supported essay with examples of evidence

3 points – Somewhat supported essay with examples of evidence

2 points – Provided a few examples to support essay

1 point – Did not provide enough examples to support essay

0 points – Provided no examples to support essay

Impact of Essay

5 points – Essay was outstanding and made the reader feel invested in the student’s essay

4 points – Essay was good and the reader felt connected to the student’s essay

3 points – Essay was okay and the reader understood what the student was trying to express

2 points – Essay had a point and the reader didn’t lose interest while reading the essay

1 point – Essay was poor and the reader had to work to engage with the essay

0 points – Essay was disjointed and the reader was unable to connect with the essay

Up to 2 bonus points were also given for applications that required supplemental information, but I’ve omitted those criteria for brevity.

After submitting my scores, I wondered how my scores compared to those of other parents. Because I was the first volunteer in my group to complete my assignment I did not have visibility into how the other 2 parents scored the students’ applications. However, I was able to externally validate my scores against those of various large language models (LLMs).

METHODS

There are too many LLMs to count nowadays, so I consulted the 7 that I was most familiar with, and I’ve listed the most probable models that each one is likely to have used as of the time of this writing. Some LLMs are more transparent with the identification and versioning of their free and paid models. For all 7 models, I used the free tier.

• ChatGPT: Default model: GPT-5.2 Instant; Fallback model: GPT-5.2 Mini or similar lightweight version if you exceed limits
• Claude: Sonnet 4.6
• Copilot: Copilot model, built by Microsoft
• DeepSeek: DeepSeek-V3.2
• Gemini: Gemini 3
• Grok: Grok 4.20 beta, Auto (Fast or Expert)
• Perplexity: model not shown or configurable on free plan

I used the exact same prompt for all 7 LLMs and all 4 students:

You are a parent of a high school student who has volunteered to evaluate scholarship applications. Students who apply for a scholarship under the category of SCHOOL SERVICE are given the following essay prompt: “What contributions have you made to our high school as someone who serves this community?” Students who apply for a scholarship under the category of LEADERSHIP are given the following essay prompt: “Would others consider you a leader and why?” OR “What is your definition of a leader and how do you embody those characteristics?”

The grading rubric is provided in the attached “Essay Scoring Guidelines.pdf” file. Provide scores as whole numbers for the following dimensions in accordance with the scoring guidelines:

1. Followed Directions (0-2 points)
2. Answered Essay Prompt (0-3 points)
3. Well-Written and Use of Good Grammar (0-5 points)
4. Provided Examples of Supporting Evidence (0-5 points)
5. Impact of Essay (0-5 points)

Ignore the “Bonus Points” dimension in the scoring guidelines because the scoring of that dimension may involve evaluation of photos or videos. The student’s essay is attached. Provide the score for each of the 5 dimensions along with a brief justification for each score.

For each model, I pasted the prompt and attached the essay scoring guidelines in a PDF file along with a PDF file the essay for student 1. I continued using the same chat thread, so I only attached the PDF files of the essays for students 2-4, as re-attaching the scoring guidelines repeatedly for each student would have been redundant. For privacy reasons, I have de-identified the student names and am not sharing the actual student essays.

RESULTS

My ratings, along with those of the 7 LLMs, are as follows (click the image to enlarge):

Although the LLMs did provide brief justifications for their scores, I’ve included only the numeric results but could easily furnish the complete LLMs responses upon request.

Overall, there was general agreement between my ratings and the average ratings from the 7 LLMs.   In terms of rank order, I gave the highest scores to Student 1 (19 points), followed by Student 4 (17), Student 2 (15), and Student 3 (12). Using the average of all 7 LLMs, the highest score went to Student 1 (19.7), followed by a 2-way tie between Students 2 and 4 (19.1), and then Student 3 (15.4). In other words, the LLMs agreed with my ratings for the best and worst applications, although they did not draw a distinction between the two applications in the middle of the pack.

Across the board, I was equally or more critical of the essays than the LLMs, as the LLMs generally gave the same or higher scores in each of the 5 dimensions of the grading rubric. Upon examining the total number of points allocated across LLMs, the 3 most “lenient” graders were Grok (78 total points awarded), Perplexity (77), and Copilot (76), while the “strictest” graders were Claude (69), ChatGPT (70), and Gemini (70).

DISCUSSION

All 7 LLMs were up to the task of grading the essays in accordance with the grading rubric. I considered the possibility that some LLMs might not completely follow directions, but all of them adhered precisely to the grading criteria and listed scores that were concordant with the criteria. Some LLMs even tallied up the total scores for each student even though I did not specifically request it in my prompt, and when they did so, they performed addition without any errors.

There are several possible explanations for the differences between my ratings and the LLM ratings. First, it is possible that I’m a tough grader. I went into this activity thinking that these were all brilliant students, and it would not be helpful if all the students clustered around near-perfect scores. In fact, this is exactly the outcome that was observed with the LLMs, as students 2 and 4 were deadlocked in a tie. Second, it is possible that the LLMs were lenient graders. After all, sycophancy in LLMs has been well-documented and researched, and many companies have made concerted efforts to tone down the level of sycophancy as they introduced new versions of their models.

This experiment validates that LLMs can be used to assess the quality of written text when evaluated against a custom rubric. This is probably not surprising to many readers who have already engaged with LLMs in similar ways, including myself. However, this is the first time I’ve quantified my findings. Another key takeaway is that LLMs can be used to critically appraise a body of written text so the author has a chance to make revisions based on the feedback. In academic settings, the mere usage of LLMs is not tantamount to cheating. It’s the way in which an LLM is used that constitutes whether the LLM serves as a learning aid or if it is used to cheat. In work settings, I encourage professionals to take full advantage of LLMs to enhance learning, spark creativity, and optimize productivity. As long as LLMs are used in a way that they do not substitute critical thinking, I think we have a lot to gain.

Chicken Al Pastor and Oxford Commas

I was driving my wife home after she had a medical procedure, and she asked me to buy her some food from Chipotle. I listened with apprehension as she rattled off a litany of food items and ingredient customizations, as I knew there would be no way that I’d get all the details right. You see, my wife has very particular preferences when it comes to food. So to ensure that I had the best chance of getting her order correct, I asked her to text the instructions to me. She initially refused, saying that I make no effort to remember her preferences. I said that if I have to remember more than 2 or 3 things about her order, I will screw it up and she will be upset. Besides, I was driving and trying to find the restaurant so wasn’t able to pay enough attention to commit her customizations to memory. So she relented and sent me the following text message (verbatim and therefore in quotes):

“Chicken Al pastor, brown black beans corn, green sauce and salsa on side”

And while I was still driving, she verbally told me to take her phone and redeem an offer for free queso by scanning a QR code provided in the app and also scan her rewards number so she could earn points. I felt that I could remember those last 2 instructions because they were the last things she mentioned, and all the other details were in the text message. I had never heard of chicken al pastor, nor did I know that Chipotle had that on their menu. It turns out that it is a time-limited offer. Note that the link may not work when the offer expires, but here’s a screenshot from that site.

After struggling to find parking, my wife stayed in the car while I entered the store and read my wife’s text message to the server. I was asked, “Burrito or bowl?” to which I requested a bowl which my wife usually gets (BTW, I received no credit for knowing the answer to this question). I also had the intuition to know that “brown black beans” meant “brown rice and black beans” despite the instructions being technically incomplete/illogical (no credit for that one either). After carefully crafting my wife’s gourmet meal, I scanned the QR code for the free queso offer and scanned the QR code for my wife’s rewards program and paid. Mission accomplished, or so I thought (foreshadowing).

When we got home, my wife asked where the green sauce was. I told her that I saw them put the green sauce in the bowl. She complained that she wanted the green sauce on the side, NOT IN THE BOWL.

----- Begin Side Conversation About Oxford Comma -----

One could argue that there was no Oxford comma in her text message, so it should have been clear that both the green sauce and salsa needed to be put on the side. However, if you read the entire text message, the punctuation is wildly inconsistent, so no reasonable person could definitively conclude that the absence of an Oxford comma necessarily meant that the green sauce should have been put on the side. Plus, I am pretty sure that my wife does not know what an Oxford comma is.

----- End Side Conversation About Oxford Comma -----

Anyway, the situation was quite upsetting to her, as she continued to complain that I never try to understand her. I found the situation to be somewhat amusing actually because not only did I anticipate that this would happen, I also called it out and tried to prevent it from happening, and it happened anyway. It’s not that I don’t try to understand my wife as a person, I just have low tolerance for complexity when it comes to fast food, so I try to shift the burden of perfecting an order back on her, and therefore I think she is partially correct on that criticism of me. Also, I think I have been conditioned to just accept that whatever I do, it will be wrong, and I will be blamed anyway.

When I order food, I’ll usually accept whatever normally comes with the dish, or in the case of a build-your-own dish scenario, I’ll just have everything. Honestly I don’t really care that much if I get white or black rice, brown or black beans, or green or red salsa. I certainly don’t need things put on the side, just dump everything in and save a plastic container from taking up space in landfills. Besides, I will eventually mix it all together and everything will come out the other end looking the same regardless of how it was prepared. And if someone orders food for me, I will say “thank you” and happily eat the food. No complaints, no drama.

I am not saying that people should not have detailed food preferences. I just think they should not impose their expectations on others and get upset when people fall short of those expectations. Also, a clearer text message such as this one could have prevented the snafu:

“Chicken al pastor in bowl, brown rice, black beans, corn, green sauce on side, salsa on side. In Chipotle app, redeem offer for free queso and scan rewards code.”

It is specific and understandable, and I just demonstrated how an Oxford comma in combination with other clear communication could have saved the day. Oh what could have been!

3D Printing Without Wi-Fi

Today I was unable to send a print job wirelessly from my Mac to my Bambu Lab A1 3D printer because our Spectrum internet service went down.

I am accustomed to sending print jobs wirelessly to my 3D printer, in fact I have never done it any other way. Because I can turn my iPhone into a hotspot with my Visible Wireless cellular plan, I connected both my laptop and 3D printer to my hotspot. The connection was slow partly because I have the basic plan with 5 Mbps hotspot speeds but also because my 3D printer is located on the first floor where cellular reception is somewhat spotty. It is good enough for phone calls but not so great when it comes to transmitting larger amounts of data.

I sliced my model in Bambu Studio as I normally do. I then sent the print job which normally occurs in 2 phases. First, Bambu Studio uploads the print job from my laptop to Bambu Lab’s cloud service. Second, it downloads the print job from the Bambu Lab cloud to the 3D printer. It slowly but successfully uploaded the 4.1 MB print job to the cloud. However, the 3D printer struggled for a while to download the print job from the cloud and eventually failed.

Therefore, I reverted to the tried and true local printing method via microSD card which bypasses the internet. After slicing my model in Bambu Studio, instead of sending the print job via the cloud, I chose the “Export plate sliced file” option. From there, a “Save sliced file as:” dialog box allowed me to save a .gcode.3mf file. I placed the .gcode.3mf file in the root directory of the microSD card that came with my Bambu Lab A1 3D printer and powered up the printer. After staring up, I pressed the “Print Files” option on the home screen and selected my .gcode.3mf file. From there, I was able to toggle options for AMS, dynamic flow calibration, and bed leveling, just as I would have done when sending a print job from Bambu Studio via cloud printing. It worked like a charm.

With my first 3D printer, a Creality Ender 3 V2 Neo, I printed exclusively via microSD card because it did not offer a wireless option (at least not natively). Although printing via microSD card is not complex, it certainly is more convenient for me to send print jobs wirelessly than to transfer my microSD card between my computer (2nd floor) and 3D printer (1st floor). Some folks have concerns about privacy when sending print jobs through Bambu Lab cloud services, but I have no such concerns because all my prints are for fun and entertainment, and I have nothing to hide. I like the convenience of cloud printing and will appreciate it even more after my Spectrum internet service is restored!

3D Model Figurine Generators

I started 3D printing as a hobby in May 2023. At the time, most of my 3D prints were of models that other people created and uploaded to free online repositories such as the ones I’ve described here. I then took the next logical step of learning a CAD application called Tinkercad to create my own simple models. For some specific use cases, I’ve experimented with 3D modeling streets and terrain. Generative artificial intelligence exploded onto the scene in recent years, and now there are many websites that allow users to upload a photo and automatically generate a 3D figurine without knowing anything about mesh modeling of curved surfaces. In this post, I compare 2 free 3D figurine generators: PrintU by Bambu Lab and FanForge by Creality.

As depicted in the image at the top, I uploaded the same photo to PrintU and FanForge and generated 3D models. Both PrintU and FanForge had relatively easy to understand wizard interfaces, and both websites offered variations for how to generate the 3D models. I generated 3 variations in each application, and screenshots of the 3D models are presented below.

As you can see, the 3 models generated by PrintU were far more realistic than the ones generated by FanForge. In fact, the FanForge models did not even remotely match the facial features that were in the uploaded photo. The FanForge models were more “artistic” which could partially explain their deviation from reality, but if the starting point is a bust photo, I have an expectation that the resultant model should bear some resemblance.

I’d have to experiment with additional models generated by different photos before drawing more definitive conclusions, but my initial impression is that PrintU is the clear winner in this head to head comparison.

Watching UFC Became More Affordable

I have been a fan of mixed martial arts for more than a decade, and I think most people would agree that the most compelling fights are in the UFC. When UFC launched in 1993, broadcasting was exclusively via pay-per-view (PPV). In 2005, UFC partnered with Spike TV which marked their transition from a fringe sport into mainstream entertainment. In 2011, UFC signed its first major network TV deal with FOX Sports.

In 2019, the UFC signed a deal with ESPN. Fans could watch UFC Fight Night events on ESPN+ (available by subscription for $11.99/month or $119.99/year in 2025), whereas numbered UFC events (featuring championship fights and more popular fighters) were all PPV events ($79.99 per event in 2025, on top of the cost of and ESPN+ subscriptions). In other words, it was very expensive to be a UFC fan.

UFC’s transition to Paramount+ in 2026 was a big deal for the UFC and its fans. The most obvious change was the abandonment of the PPV model. Now a subscription to Paramount+ gives viewers access to both Fight Nights and numbered events. Paramount+ Essential is ad supported and costs $8.99/month or $89.99/year. Paramount+ Premium without ads costs $13.99/month or $139.99/year. For a hardcore UFC fan who watches every event, the cost of watching UFC on Paramount+ is roughly one tenth the cost of ESPN+. Hooray!

But how does this make any sense from a business perspective? I believe that the strategic play from Paramount+ is to change the model from price-gouging devoted UFC fans to adding compelling live sports content to its streaming platform. The net effect I think Paramount+ is aiming for is to increase the number of subscribers and to increase retention. On the latter point, when I subscribe to a streaming service, there are a limited number of movies and shows that I enjoy watching, so I am likely to eventually cancel my subscription. However, given that UFC has live events almost on a weekly basis, I will continue to tune in and am more likely to renew my subscription.

I subscribed to the ad-supported Paramount+ Essential plan and really enjoyed watching the first 2 events of 2026, UFC 324 and UFC 325 (which by the way is the first time that I ever recall there being back-to-back numbered UFC events). It was great to watch these numbered UFC events without the PPV price tag. One complaint I had about the broadcasting of UFC 324 was that they didn’t do a great job with the timing of the advertisements. I had expected ads with my Paramount+ Essential plan, but sometimes they would run ads in between rounds when I would prefer to hear what is being discussed in the fighters’ corners. This very issue was discussed in the UFC 324 post-fight press conference with UFC CEO Dana White at the 5:06 mark. I felt that they did a better job in UFC 325 in broadcasting the corner work, so it feels to me that Dana White and Paramount+ had targeted that as an area of improvement, and I expect that this will be a permanent change.

I previously wrote about how the UFC excels not only as a mixed martial arts company but also as a production company. With Paramount+, the UFC continues to provide undisputedly fantastic content and does so at an affordable price. Under Paramount+, I look forward to seeing further growth of the UFC brand as its fights become more accessible to general audiences. By extension, I hope this also results in the growth of the sport of mixed martial arts as a whole.

3D Printing in the Crosshairs in Washington State

Washington State recently introduced House bills HB 2320 and HB 2321 that aim to restrict the use of 3D printing technology in order to prevent the illegal manufacturing of firearms. While the stated goal of improving public safety is understandable, I feel that these bills are misdirected at the maker community and are a setup for a whole series of unintended consequences. Instead of offering a thoughtful solution, the proposals rely on simplistic and heavy-handed restrictions that may be difficult to implement, costly and impractical to enforce, and largely ineffective at stopping illegal activity.

HB 2320 is entitled “Concerning the regulation of firearm manufacturing” and focuses on expanding regulations around firearm manufacturing, particularly when digital tools like 3D printers, CNC machines, and downloadable design files are used. The bill broadens existing definitions of firearm manufacturing to explicitly include digital methods and makes it illegal to create certain gun parts or firearms without proper licensing. In practice, this would criminalize a wide range of hobbyist activities, even when no harmful intent exists. Many makers use 3D printers for educational projects, prototyping, and mechanical experimentation, and the bill risks sweeping these legitimate uses into a legal gray area. The concern is that well-meaning individuals could face legal consequences simply for owning tools or files that resemble firearm components. Taking a closer look at section 8 on page 31, the proposed new law under HB 2320 would state, among other things:

To my knowledge, there does not exist a 3D printer manufacturer that has the “primary or intended function” of manufacturing firearms. 3D printers can be instructed to make many things, of which firearms are a very small subset. To use an analogy, it would be similar to assuming that car manufacturers make cars that have the primary or intended function of taking you to the gun store, so maybe we should ban cars. This bill represents a gross misunderstanding of 3D printing technology.

HB 2321 is entitled “Requiring three-dimensional printers be equipped with certain blocking technologies” and goes even further by requiring that all 3D printers sold or transferred in Washington be equipped with software that can detect and block the printing of firearm parts as described in section 8 on page 6:

The bill would require manufacturers to implement “firearm blueprint detection algorithms” and prevent printers from producing restricted designs. While this may sound straightforward, the reality is far more complicated. 3D printer files can be easily modified, disguised, or broken into smaller pieces, making detection unreliable. In addition, many 3D printers operate offline or use open-source software, making enforcement nearly impossible, unless all “compliant” 3D printers are forced to use a restricted set of software and firmware which would essentially stifle innovation and progress with 3D printing.

As discussed in this Reddit post from the Seattle community, these rules could also harm small businesses, educators, researchers, and hobbyists who rely on open and flexible printer systems for innovation.

The enforcement challenges of both bills are significant. Policing digital files and machine firmware requires advanced technical oversight that state agencies may not be equipped to handle. Manufacturers would face higher production costs, which would likely be passed on to consumers. Meanwhile, individuals intent on producing illegal firearms could simply bypass these systems by modifying firmware, using older printers, or obtaining equipment from outside the state. This means that law-abiding users would bear the burden of regulation, while bad actors could easily evade it and undermine the effectiveness of the legislation.

Ultimately, these bills reflect a misunderstanding of the 3D printing community and the technology itself. I honestly don’t know how to curb illegal firearm manufacturing or how to address the epidemic of gun violence in the United States, but I highly doubt that HB 2320 and HB 2321 would curb the actions of bad actors or help improve public safety. Instead, they will create unnecessary barriers to innovation and ruin the careers and hobbies of peaceful members of the maker community.

Apple Wallet IDs

For the past few years, I’ve been using the Wallet app on my iPhone to make electronic credit card payments and scan my airline boarding passes. I now have another use case: presenting my ID while going through airport security. You see, Wallet now supports ID verification for use at TSA checkpoints or to show proof of identity or age at selected businesses, venues, websites, or apps.

You can add your driver license/state ID and/or your passport to create an ID in your Wallet app. Visit https://learn.wallet.apple/id for step-by-step instructions. Scanning my passport resulted in the creation of a “Digital ID” card in my wallet app. The verification process was very fast, as I received an email a few minutes later to notify me that my Digital ID was ready for use. I used the same steps to scan my California driver license which resulted in the creation of a “California State ID” card in my wallet app. The verification process was also fast, as I received an email from the California Department of Motor Vehicles that my mobile driver’s license (mDL) was successfully added to my Apple Wallet. I had previously enrolled in the mDL pilot so I’m not sure if that helped expedite my verification.

For my ensuing business trip, I tested the use of my Digital ID card at a TSA checkpoint. I simply opened my Digital ID card, double-clicked the side button, and scanned my iPhone on the TSA reader. See Apple’s Use your Digital ID in Apple Wallet support page for more information.

Many years ago I had accidentally departed on a business trip without my physical wallet which contained my driver license and credit cards. To get to the airport, I had become so accustomed to hailing an Uber or Lyft ride on my phone that I completely forgot to bring my wallet. I didn’t realize my blunder until I arrived at the airport, and I didn’t have enough time to return home to fetch my wallet and get back to the airport in time to catch my flight, However, I was able to board my flight after going through some extensive TSA screening procedures, and I was told that it helped that I had TSA Precheck, otherwise the screening would have taken longer. I later called my wife who sent my wallet to me via overnight express so I could use my ID and credit card for the remainder of my business trip.

As I reflect on that experience, I realize that if I leave home without my wallet for a business trip today, I will probably be just fine without my physical wallet because I have everything I need on my iPhone. With Apple Wallet, I can use my Digital ID to go through TSA checkpoints, I can scan my boarding passes to board planes, and I can use all my credit cards to make payments everywhere electronic payments are accepted. With my rideshare apps (mostly Uber and Lyft but now also Waymo in some cities), I can get to anywhere I need to go. I’d only be in a jam if I need to make a cash payment, but honestly I don’t remember the last time I had to use cash on a business trip, other than leaving a tip in a hotel. That being said, I still plan to travel with my physical wallet for the time being, just as a precaution.

Los Angeles Neighborhoods

If you live in or have visited the Los Angeles area, you may know that Los Angeles is a big place, and sometimes people like to specify neighborhoods in Los Angeles where they live, work, or play. Did you know that if you have an address, you can look up the neighborhood in which it resides? The LA Times Neighborhood Boundaries website allows you to either search for an address or zoom/pan to a location. Mousing over the location reveals the neighborhood for that location.

You may have also noticed that sometimes there are areas adjacent to Los Angeles neighborhoods that are actually cities. For example, Hollywood is a neighborhood in Los Angeles, but West Hollywood is an incorporated city within Los Angeles, immediately west of Hollywood as the city name implies. Similarly Beverly Hills is a city entirely contained within the city of Los Angeles, but immediately east of Beverly Hills is the neighborhood of Beverly Grove which does not have a city designation.

The difference between a neighborhood and a city is that a city is an incorporated municipality. That means a city is legally created under state law, has its own local government, can pass its own laws, and has its own official United States Postal Service (USPS) city name. That being said, USPS does generally recognize neighborhoods as aliases for the official city names. For example, I used to live in the Hollywood neighborhood. Although the proper USPS city designation is Los Angeles, mail that was sent to me using “Hollywood” as the city name would still reach my mailbox—probably aided by the use of the proper zip code.

Californians: Protect Your Privacy for Free!

According to the California Privacy Protection Agency, a data broker is a business that gathers and sells consumer information that the consumer didn’t give to them directly. This can include personal information such as a person’s social security number, geolocation, email addresses, health-related information, and other sensitive information. Data brokers often collect information from businesses you interact with directly (e.g., music or video streaming services, grocery stores, and other businesses). More information about data brokers is provided here.

In response to the Delete Act, a new Delete Request and Opt-out Platform (DROP) service was created for California residents. By creating a profile and submitting a DROP request, Californians can protect their privacy. DROP limits the information that data brokers collect and sell about you, and it is completely free. When your data stops getting sold, you’ll presumably also have fewer unwanted texts, calls, and emails. You should in theory also have better security through a lower risk of identity theft, fraud, AI impersonations, and data leaking/hacking incidents.

DROP launched on January 1, 2026, and data brokers will begin processing requests on August 1, 2026. Data brokers will then be required to delete data every 45 days.

If data brokers do not comply with the law, they may face fines as described here. I’ve registered myself and will check my DROP status after data brokers begin processing requests in August. I plan to get my family members to register as well.

American Airlines Free Inflight Wi-Fi

Hello from cruising altitude! I’ve flown American Airlines for many years, and they just started offering free in-flight Wi-Fi to AAdvantage members. An email that I received today states:

“Free Wi-Fi is available on today's flight: If you're an AAdvantage® member, you'll be one of the first to try free high-speed Wi-Fi on your upcoming flight. Once onboard, go to aainflight.com and use your AAdvantage® login.”

I would not consider the internet speed to be a “high-speed” connection, but it’s hard to be upset about free inflight Wi-Fi.

Normally I do not pay for inflight Wi-Fi, but I’m sure I’ll take full AAdvantage of it when available. I did not see any information about which flights would be offering free Wi-Fi, but given that I normally fly between big cities, I hope I will have it often.

Hospital Website Wiretapping Settlements

The Meta Pixel tracker was exposed many years ago for its nefarious practices of obtaining health-related information about patients from hospital websites, and I wrote about it here on June 25, 2022.

However, Meta/Facebook was not the only bad actor. Google and many other companies had also been known to “wiretap” hospital websites to collect protected health information (PHI). As a result, the Office of Civil Rights issued statements on the matter, along with warnings to hospital systems about the use of these online tracking technologies. I wrote about it here on July 23, 2023.

To summarize the legal issues, there is a healthcare-specific law called the Health Insurance Portability and Accountability Act (HIPAA) which prohibits healthcare providers from sharing PHI with third parties. This law appears to have been repeated violated by the implementation of tracking technologies on hospital websites.

Yesterday I received a Notice of Proposed Class Action Settlement in which Kaiser Foundation Health Plan, Inc. reached a proposed settlement of claims related to their disclosure of patients’ confidential personal information due to third-party software code. The Consolidated Master Class Action Complaint specifically mentions Quantum Metric, Twitter, Adobe, Microsoft Bing, and Google as “Third Party Wiretappers” on paragraph 4 (page 2) and states that it is not limited to just these companies.

In paragraph 261 (page 74), it specifically mentions that Meta/Facebook and Google Analytics have been implicated in this specific tracking activity of personal health information.

There have been other lawsuits against other healthcare delivery organizations and resultant settlements. I used ChatGPT to help me identify some of them which include University of Tennessee Medical Center & Margaret Mary Community Hospital, Pomona Valley Hospital Medical Center, Louisiana Children’s Medical Center, Reid Health, MarinHealth, and Eisenhower Medical Center to name some lawsuits in which Meta Pixel was specifically implicated. I have not done comprehensive research, but I suspect that there are many others in which other third party wiretappers are named. If you were a Kaiser patient between November 2017 and May 2024, and if you haven’t received notification about the Kaiser Privacy Breach Settlement, visit https://kaiserprivacysettlement.com to learn more.