Nephi Walton

Clinical Geneticist, Bioinformaticist, Recovering Programmer, Writer, Science Lover, Anatomical Oddities, Nerd, Dad, @WakeForest, Opinions are my own

[Nephi Walton @geneticks]

You’ve got a great product, I use both Loveable and Grok, (and Claude). My 12 year old son won a competition using Loveable. It really opened the door to novice developers. Seeing mutual respect between competitors makes me even more supportive of both platforms. I love the way competition fuels innnovation. That said, it’s moving so fast at this point, keeping up to date is becoming a bit exhausting.

[Nephi Walton @geneticks]

I’m excited to try this. Installing now

[Claude @claudeai]

Introducing Claude Science, a new app designed with every stage of research in mind.

Artifacts traced to their code, environments managed on demand, and 60+ optional scientific databases that you can connect.

Available now in beta.

[Claude @claudeai]

Introducing Claude Science, a new app designed with every stage of research in mind.

Artifacts traced to their code, environments managed on demand, and 60+ optional scientific databases that you can connect.

Available now in beta.

[Nephi Walton @geneticks]

@midjourney As a bioinformaticist and clinical geneticist with a wife whose a sonographer, there is huge research and discovery potential here. Are you open/interested to research collaborations particularly in the AI/genomics space?

[Nephi Walton @geneticks]

This is really quite remarkable in terms of understanding the origin of several human genetic disorders. https://t.co/PZNkTQHRPU

[Nephi Walton @geneticks]

I should add, that I was so impressed that I ended up buying one two weeks later and we used it for the project. FSD has been great for me too. I actually just relax and communicate with Grok about genetic conditions I am seeing as the car is driving me to work. Driving is much less painful than it used to be. My thought was that if they approved her to drive with FSD that I would hand over the keys, but she might have to get her own Tesla (with some help).

[Nephi Walton @geneticks]

I think the use of FSD for people with disabilities has tremendous potential. Nearly 3 years ago, we discovered my daughter had epilepsy, she has not had a tonic clonic seizure since but continues to have small myoclonic jerks of her hands once or twice a day that last less than 2 seconds. These can be embarrassing and she drops things occasionally when this happens. As she is 15 and turning 16 soon, the biggest questions she has had at each appointment is "when can I drive". After an appointment in December she was very discouraged with no answer as to when she could drive. To lighten her mood I took her to a Tesla dealership and we test drove FSD (technically the car drove the whole time). This made her extremely happy as she could see that some day regardless of her seizure control she would be able to drive. This inspired her entry into the Presidential AI competition for which her team won in the state of Utah. She thinks, and I agree, that FSD could be prescribed by clinicians to help people with epilepsy like hers gain independence today, but could be even better if other features were integrated into Tesla’s internal camera system.
@Tesla
and
@elonmusk
would love to have your thoughts on this. She is going to continue working on this even if she does not win regionals, because it is extremely important to her.

[Nephi Walton @geneticks]

@EdwardLawr82368 @XFreeze Yes, my daughter actually brought that up in one of her presentations

[Nephi Walton @geneticks]

Nearly 3 years ago, we discovered my daughter had epilepsy, she has not had a tonic clonic seizure since but continues to have small myoclonic jerks of her hands once or twice a day that last less than 2 seconds. These can be embarrassing and she drops things occasionally when this happens. As she is 15 and turning 16 soon, the biggest questions she has had at each appointment is "when can I drive". After an appointment in December she was very discouraged with no answer as to when she could drive. To lighten her mood I took her to a Tesla dealership and we test drove FSD (technically the car drove the whole time). This made her extremely happy as she could see that some day regardless of her seizure control she would be able to drive. This inspired her entry into the Presidential AI competition for which her team won in the state of Utah. She thinks, and I agree, that FSD could be prescribed by clinicians to help people with epilepsy like hers gain independence today, but could be even better if other features were integrated into Tesla’s internal camera system. @Tesla and @elonmusk would love to have your thoughts on this. She is going to continue working on this even if she does not win regionals, because it is extremely important to her.

[Nephi Walton @geneticks]

My daughter animating her Easter eggs with Grok Imagine

[Nephi Walton @geneticks]

Witnesses Testify on FDA Regulations and Innovation https://t.co/t2HFBZyAl5

As a clinician who cares for patients with these disorders, I am deeply troubled by recent FDA actions. While there has been some encouraging news, it has not translated into meaningful progress. Many of these patients—particularly those with spinocerebellar ataxia—are losing function every day that they will never regain, even as therapies with demonstrated positive effects exist but remain out of reach. This is not a political issue. It is profoundly frustrating to watch patients deteriorate rapidly when a potentially beneficial intervention exists, yet incomprehensible regulatory barriers continue to delay access.

[Nephi Walton @geneticks]

Grok is very good. I’m a clinical geneticist and I’m seeing more and more patients come up with correct diagnoses using LLM’s. In several cases they are trying to get genetic testing from a specialist who brushes them off. For those, it’s making them lose faith in the medical system when they feel they have more information than their provider. We have to embrace AI as clinicians or risk losing the trust of our patients. For my field it’s impossible for me to memorize the 10,000 plus diagnoses I see. For a computer this is trivial. Physicians still have value exceeding LLM’s for patient care, but we need to recognize with humility where AI beats us and utilize it rather than shun it, anyone who doesn’t will become obsolete.

[Nephi Walton @geneticks]

Great perspective on AI in education.

[Andrej Karpathy @karpathy]

A number of people are talking about implications of AI to schools. I spoke about some of my thoughts to a school board earlier, some highlights:

1. You will never be able to detect the use of AI in homework. Full stop. All "detectors" of AI imo don't really work, can be defeated in various ways, and are in principle doomed to fail. You have to assume that any work done outside classroom has used AI.
2. Therefore, the majority of grading has to shift to in-class work (instead of at-home assignments), in settings where teachers can physically monitor students. The students remain motivated to learn how to solve problems without AI because they know they will be evaluated without it in class later.
3. We want students to be able to use AI, it is here to stay and it is extremely powerful, but we also don't want students to be naked in the world without it. Using the calculator as an example of a historically disruptive technology, school teaches you how to do all the basic math & arithmetic so that you can in principle do it by hand, even if calculators are pervasive and greatly speed up work in practical settings. In addition, you understand what it's doing for you, so should it give you a wrong answer (e.g. you mistyped "prompt"), you should be able to notice it, gut check it, verify it in some other way, etc. The verification ability is especially important in the case of AI, which is presently a lot more fallible in a great variety of ways compared to calculators.
4. A lot of the evaluation settings remain at teacher's discretion and involve a creative design space of no tools, cheatsheets, open book, provided AI responses, direct internet/AI access, etc.

TLDR the goal is that the students are proficient in the use of AI, but can also exist without it, and imo the only way to get there is to flip classes around and move the majority of testing to in class settings.

[Nephi Walton @geneticks]

The next innovation that will really impress me is adding the ability to add future alarms. Something users have asked for since 2013 and @Apple has ignored. Two people I know have switched to android just because of this and I’m seriously considering it. Android has done this for a long time.

[Nephi Walton @geneticks]

I wouldn’t brag about this @Uber AI support is the worst I’ve encountered to date. Just came out of an AI conference, experienced Ubers ai chatbot firsthand. After a ride never showed up and still charged me I spent 30 minutes dealing with the worst AI chatbot I’ve ever encountered and switched to @lyft. It’s bad examples like this that will give people bad impressions of the true potential of ai. Came out of this conference with so much excitement for ai then get slapped in the face with this horrible real world example. AI is not magic, still have some work to do, at least @uber does.

My expectation is that it shouldn’t take 30 minutes to communicate that my driver never showed up and I never got in a ride I’m being charged for. I don’t think that’s too high of an expectation

[Firebolt @FireboltHQ]

This time on the Data Engineering Show: Paarth Chothani from @Uber shares how they're innovating on-call support with AI 🤖

Highlights:
🔹How Uber built Genie to tackle the universal pain of on-call support
🔹The brilliant progression from basic documentation assistant to automated problem-solver
🔹Real examples of AI agents working together to resolve complex issues

Tune in: https://t.co/UcMT6kDxwl
Spotify: https://t.co/0HiPe4Njyh

#AI #dataengineering #podcast

[Nephi Walton @geneticks]

@FireboltHQ @Uber @Uber AI support is the worst I’ve encountered to date. Just came out of an AI conference, experienced Ubers ai chatbot firsthand. After a ride never showed up and still charged me I spent 30 minutes dealing with the worst AI chatbot I’ve ever encountered and switched to @lyft

[Nephi Walton @geneticks]

Great panel on AI and the patient experience. Discussed using AI to make visits more efficient to address physician burnout. I have a feeling that health systems will pay for AI to increase efficiency and then increase patient volumes, therefore sustaining burnout. Thoughts?