Wednesday, August 23, 2023

New Kid

 Claude 2, like Google Bard, is free to use, though it does require registration with your full name.  It claims a more up-to-date body of training data.

I thought this new Large Language Model handled my initial test question quite nimbly:

* * *

Compare the prevalence of conspiracy theories in U.S. political discourse to that of other democracies.

There are a few key points to consider when comparing conspiracy theories in US political discourse versus other democracies:

The US has a long history of conspiracy theories intertwined with politics, going back to the anti-Masonic movements of the 19th century and fears of Catholic infiltration. This provides fertile ground for new conspiracy narratives to take hold.

American exceptionalism and distrust of government power feeds into conspiratorial thinking - the idea that "they" are out to undermine American greatness. This is less prevalent in cultures with more collective/communal values.

The two-party polarization in the US exacerbates an "us versus them" mentality. Hyper-partisanship makes each side prone to believing negative conspiracy theories about their opposition.

The US system of checks and balances, separation of powers, and freedom of speech makes it harder to clamp down on misinformation. Conspiracy theories spread readily on social media.

Right-wing media ecosystem in the US propagates conspiracy narratives, for example about Obama's birthplace or the 2020 election. This has less equivalence in other democracies.

Comparatively, countries like the UK or Australia have had fewer high-profile political conspiracy theories get broad traction. However, the internet age is impacting all open democracies.

So in summary, there are cultural and systemic factors that provide fertile ground for conspiracy theories to take hold in the US political mainstream more so than peer nations. However, other democracies are impacted too by misinformation trends in the digital age.

Friday, August 18, 2023

Automating Health Care Provision

 Dealing with COVID has stimulated some thoughts about the current state of health care and where it is likely heading under the influence of Artificial Intelligence and Large Language Models.

The day after we arrived home from our Portland adventure I started an effort to see what treatment options were available.  I found out that COVID emergency procedures had been shut down and the only option which seemed available was my regular doctor at the Presbyterian Health Services clinic.  

The phone listed under my doctor's name did not work, so I dialed the main clinic number and got put on endless holding music.  I then sent the following message to the care team:

I tested positive on Tuesday evening for COVID.  I have some congestion and fatigue.  My temperature is normal.  I would like a recommendation for treatment.  

Mike Connealy

Two days later I got a reply:

Hello Mr. Connealy,  

 We received your MyChart message and due to the symptoms you have reported, your message has been sent to a Team Registered Nurse (RN). 

 We will make our best efforts to address your concerns and determine the next steps in your care within 4 business hours...

Fortunately, in the meantime I had come across another option on the Presbyterian web portal called Online Visits.  I clicked the link, and up popped a questionnaire asking for symptoms and health history which I completed in about ten minutes.  An hour later I got a thorough emailed summary of the "online visit":

Good morning— I’m prescribing Paxlovid for you, but I wanted discuss the interaction that it will have with your levothyroxine... 

Diagnosis - COVID-19 infection

My name is ALEXANDRA CARTER, MD, and I'm a healthcare provider at Presbyterian Healthcare Services. I've reviewed your interview. Based on your responses and recent positive COVID-19 test, I see that your symptoms are caused by a COVID-19 infection.

Since you tested positive on a self-test, you don't need another test. A positive self-test means that the COVID-19 virus was detected, and you very likely have a COVID-19 infection.

Most people with COVID-19 have mild to moderate symptoms and can rest at home until they get better...

As it turned out, my regular nearby Walgreens Pharmacy did not have Paxlovid in stock, but they sent me to another branch up the road where I got the prescription at no cost.

Margaret went through a similar process with her doctor, so we also used the online visit feature to get her an immediate response and prescription.  Both her doctor and my doctor's care team expressed irritation that we had sped up the diagnosis and prescription process by doing an end-run around their usual bureaucratic procedure.

So, it seems most primary care physicians are going to have to up their game with the regular inclusion of an automated process of diagnosis and treatment.  Not every condition will be as simple to deal with as our COVID experience, but progress in AI medical care provision  will surely make it relevant to treatment on a broad scale.  

The Online Visit process in this case is perhaps best characterized as AI-lite, but it was quick and effective.  While a lot of the content in the visit summary was boiler plate, it clearly took into account my responses in terms of symptoms and health history, and it was reviewed by "ALEXANDRA CARTER, MD".  (I'm assuming she is a real person.)  I wonder if reviewing AI-generated diagnoses and treatments is a full-time job and how Dr. Carter feels about doing it.

I am also wondering about how the big health care corporations like Presbyterian are going to assimilate AI, and to what ends. While speed and accuracy are likely targets, quite a lot of motivation is also likely to involve profit maintenance and enhancement. There is no guarantee that patient benefits will be on a par with bottom line considerations.

The doctor will see you now.


Thursday, August 17, 2023

Air Travel

 Avoid it if you can.

We booked our Southwest flight to Portland well ahead of our travel date.  The flight to Portland by way of Oakland featured the usual inconveniences, but was largely uneventful.

We changed our return reservation to a day earlier than originally planned.  When we got to the baggage counter at PDX we were informed that the flight had been cancelled.  A helpful baggage person got us assigned to a later flight.  That one ended up being too late to make a connection in Oakland, so we ended up with yet a different flight.  In the end we spent about eight hours sitting in Portland's airport.

Our 737 landed in Oakland at the same time our flight onto Albuquerque started boarding.  Our arrival gate and the departure gate were at opposite ends of the terminal.  Unlike Portland, there were no moving sidewalks at Oakland,  so I hobbled along at the best speed I could muster for what seemed like about half a mile.  We made it in time to get aboard, but I arrived in pretty poor shape.

Travel

Our trip to Portland went mostly as planned, with a surprise ending.  Changing planes in Oakland provided  a welcome glance at San Francisco including the Bay Bridge, the downtown and San Bruno Mountain where I had spent many hours hawk watching.

(Right click image to open in a separate tab to view full-size image.)

In Portland we stayed in an elegant apartment on the bank of the Clackamas, surrounded by the huge trees that are everywhere.


The Clackamas has about the same flow as the  Rio Grande, being only waist deep in the middle.
Lots of wildflowers and birds, including a majority of vultures.  My daughter was lucky enough to see some aerial combat between a pair of Osprey and a Bald Eagle.


My best find along the river bank was wild blackberries,  which evoke my best memories of both San Francisco and the Pacific Northwest.

The wedding which brought us to Portland included an hour's boat trip on the Willamette.  Portland is a marvelous city; I think it may have doubled in size since my last visit.  The bridges were familiar; I was impressed with the number of modern-looking skyscrapers and the endless public open spaces.

(I only got a few shots on film on this trip.  I did better on earlier trips to Portland in 2008 and 2011.)

Our visit was brought to an abrupt end when -- on the day before the wedding ceremony -- the groom tested positive for COVID.  So we moved our departure up a day, which resulted in a complicated flight home.  A day after our return to Albuquerque we both tested positive.

Tuesday, August 8, 2023

Travel Prep (Google Bard)

 Which type of locations or environments are at highest risk for COVID contagion?

The highest risk locations or environments for COVID contagion are:

Indoor locations with poor ventilation. This is because the virus can spread through respiratory droplets that linger in the air for hours. Examples of indoor locations with poor ventilation include:

Bars and restaurants

Gyms and fitness centers

Churches and other places of worship

Schools and universities

Offices

Nursing homes and other long-term care facilities

Locations where people are crowded together. This is because the virus can spread more easily when people are in close proximity to each other. Examples of crowded locations include:

Concerts and sporting events

Festivals and fairs

Public transportation

Grocery stores and other essential businesses

Locations where people are singing or shouting. This is because singing and shouting produce more respiratory droplets, which can spread the virus more easily. Examples of locations where people sing or shout include:

Choirs

Karaoke bars

Sports stadiums

Protests

Healthcare settings. This is because people who are sick with COVID-19 are more likely to be shedding the virus. Examples of healthcare settings include:

Hospitals

Urgent care clinics

Doctor's offices

Pharmacies

It is important to note that the risk of COVID contagion can vary depending on the specific location or environment. For example, a crowded indoor location with poor ventilation is likely to be riskier than a sparsely populated outdoor location with good ventilation. It is also important to consider the specific activities that are taking place in a location. For example, a choir rehearsal is likely to be riskier than a grocery store run.

Here are some tips for reducing your risk of COVID contagion in any location:

Wear a mask that fits well and covers your nose and mouth.

Stay at least 6 feet away from other people.

Avoid crowded indoor spaces.

Open windows and doors to improve ventilation.

Get vaccinated and boosted against COVID-19.

Wash your hands often with soap and water.

Avoid touching your face.

If you are sick, stay home and get tested for COVID-19.

By following these tips, you can help protect yourself and others from COVID-19.


What precautions are most effective for individuals at reducing exposure to COVID?

The most effective precautions for individuals at reducing exposure to COVID are:

Getting vaccinated and boosted against COVID-19. Vaccination and boosters are the best way to protect yourself from getting sick with COVID-19. They can also help reduce the severity of illness if you do get sick.

Wearing a mask that fits well and covers your nose and mouth. Masks can help block respiratory droplets that may contain the virus. It is important to wear a mask that fits well and covers your nose and mouth to get the most protection.

Staying at least 6 feet away from other people. The virus can spread through respiratory droplets produced when an infected person coughs, sneezes, or talks. Staying at least 6 feet away from others can help reduce your exposure to these droplets.

Avoiding crowded indoor spaces. Crowded indoor spaces are where the virus can spread more easily. If you must go to a crowded indoor space, wear a mask and stay at least 6 feet away from others.

Opening windows and doors to improve ventilation. Good ventilation can help reduce the spread of the virus. If you are in an indoor space, open windows and doors to let fresh air in.

Washing your hands often with soap and water. Washing your hands can help remove the virus from your hands. Be sure to wash your hands for at least 20 seconds.

Avoid touching your face. The virus can enter your body through your eyes, nose, and mouth. Avoid touching your face with unwashed hands.

If you are sick, stay home and get tested for COVID-19. If you are sick with symptoms of COVID-19, stay home and get tested. This will help prevent you from spreading the virus to others.

By following these precautions, you can help protect yourself and others from COVID-19.


How significant is the risk of COVID during air travel in the U.S.?

The risk of COVID during air travel in the U.S. is relatively low. This is because of the following factors:

Airplanes have very good ventilation systems. The air in an airplane cabin is constantly being circulated and filtered, which helps to remove any respiratory droplets that may contain the virus.

Passengers are required to wear masks on airplanes. This helps to block the spread of respiratory droplets.

Airlines are taking additional precautions to prevent the spread of COVID. These precautions may include requiring passengers to show proof of vaccination or a negative COVID test, and conducting temperature checks at the airport.

According to a study by the Harvard T.H. Chan School of Public Health, the risk of getting COVID on a plane is about 1 in 2,000. This is much lower than the risk of getting COVID in other indoor settings, such as bars and restaurants (1 in 100) or gyms (1 in 50).

Of course, there is no such thing as zero risk. If you are concerned about the risk of COVID during air travel, you can take additional precautions, such as:

Booking a window seat. This will help to reduce your exposure to other passengers.

Bringing your own food and drinks. This will help you to avoid contact with surfaces that may be contaminated with the virus.

Washing your hands often with soap and water. This will help to remove any virus that may be on your hands.

Avoid touching your face. The virus can enter your body through your eyes, nose, and mouth. Avoid touching your face with unwashed hands.

If you are sick with symptoms of COVID-19, you should not travel by plane. Stay home and get tested for COVID-19.

(?)


Are any major U.S. airlines currently requiring passengers to wear masks in flight?

As of August 8, 2023, no major U.S. airlines are currently requiring passengers to wear masks in flight. The Transportation Security Administration (TSA) lifted the federal mask mandate for public transportation, including airplanes, on April 18, 2022. This means that passengers are free to choose whether or not to wear a mask on airplanes.

However, some airlines may still require masks in certain circumstances. For example, JetBlue requires masks for passengers who are not fully vaccinated. And Alaska Airlines requires masks for passengers who are traveling to or from certain international destinations.

It is always a good idea to check with your airline before your flight to see if they have any specific mask requirements.

Saturday, August 5, 2023

A Small Step Forward

ALBUQUERQUE, N.M. (KRQE) – The University of New Mexico Hospital (UNMH) is the new healthcare provider for inmates at the Metropolitan Detention Center (MDC). Officials say the new partnership makes sense for providing improved care for the facility, which averages about 1,500 inmates each day..."

* * *

The transfer of responsibility for inmate health care in Albuquerque provides a window to understanding issues and problems in jails and prisons throughout the country.  

Firstly, the city has made the change because of deficiencies in care provision by the private provider, Corizon Health, in regard to routine care as well as in-custody mortality. Such poor performance is typical of private providers nation-wide.  Those private companies win contracts for prison management and service provision by unrealistically low bids.  Once in place, the companies inevitably find it impossible to achieve profitability and to meet contract requirements without corner cutting.

A frequent corner cut to preserve profits is short staffing in health care programs.  That further exacerbates general staff shortages which plague prison management, whether it is being performed by private management companies or by local and state-run institutions. Twenty-five percent shortages seem to be the rule nation-wide.  As a result of such deficits, security is frequently compromised and, in the case of health care needs, inmates often are denied needed care simply because there is no staff person available to walk them down a hall for scheduled appointments. The inevitable outcome is a failure to control chronic health conditions, as well as a higher than acceptable inmate death rate.

Underlying all of the above is the fact that there are a lot of people in jails and prisons who should not be there.  On any given day about about five hundred of the 1500 MDC inmates on entry have no fixed address -- they are the homeless.  The majority of those have been arrested for minor offenses like trespassing and even jaywalking, and mental illness and substance abuse are very likely to be contributing factors.  Even in the unlikely event that conditions are optimal, imprisonment is not therapeutic. Inmates with mental health issues are quite likely to be discharged in poorer shape than when they entered.

Most of the above is a summary of a recent fifteen minute interview on New Mexico In Focus of  reform advocate, Peter Cubra. Definitely worth a look.

Thursday, August 3, 2023

DIY an LLM

I have been playing with Google Bard, a Large Language Model (LLM) service provided by Google. The results have been impressive in some instances but less so in others, probably due in a large part to the prompts which I provided.  For instance, my questions to BARD about the U.S correctional system and prison reform produced concise, well expressed answers.  On the other hand, requests for a report on the histories of Haiti and the Dominican Republic gave me brief and superficial results, much inferior to what I found on Wikipedia on the same subjects.

Some or all of the differences I have seen in the quality of results from BARD are clearly related to my own ignorance about the subjects of Artificial Intelligence and Large Language Models.  Fortunately, simply executed searches using easily accessed services like Google Search will turn up useful entry points for learning about these subjects.  For instance, a search on the terms, "LLM Portability", turned up a good summary of LLM capabilities which in turn suggest additional lines of inquiry.

I decided as a next step to use the BARD LLM to get some quick answers about LLM operational minimums which provided me with the following:

* * *

(Me)  What is the minimum hardware required to run an LLM?


(Bard)  The minimum hardware required to run an LLM depends on the size of the model. For smaller models, you can get away with a standard desktop computer with 16GB of RAM and a decent CPU. However, for larger models, you will need more powerful hardware, such as a server with 128GB of RAM or more and a high-end GPU.

Here are some general hardware recommendations for running LLMs:

RAM: 16GB for small models, 128GB or more for large models
CPU: A decent quad-core CPU or better
GPU: A high-end GPU with at least 8GB of VRAM
Storage: A fast SSD drive
In addition to these hardware requirements, you will also need a powerful operating system, such as Linux or Windows Server.

Here are some specific examples of hardware that can be used to run LLMs:

Desktop computer: A standard desktop computer with 16GB of RAM and a decent CPU can be used to run smaller LLM models.

Server: A server with 128GB of RAM or more and a high-end GPU can be used to run larger LLM models.

Cloud computing: Cloud computing services, such as Google Cloud Platform or Amazon Web Services, can be used to run LLM models on powerful hardware.

The cost of running an LLM will vary depending on the size of the model and the hardware that is used. For smaller models, the cost can be relatively low. However, for larger models, the cost can be quite high.

* * *

(Me)  Provide some examples of results that can be obtained from a Large Language Model running on a desktop computer.


(Bard)  Sure, here are some examples of results that can be obtained from a Large Language Model running on a desktop computer:

Text generation: The model can be used to generate text, such as poems, code, scripts, musical pieces, email, letters, etc. The text generated by the model will be similar to human-written text, and it can be used for a variety of purposes, such as creative writing, marketing, and customer service.

Translation: The model can be used to translate text from one language to another. The translations produced by the model will be more accurate than those produced by traditional machine translation systems, and they will be more natural-sounding.

Question answering: The model can be used to answer questions about a wide range of topics. The model can access and process information from the real world through Google Search, and it can use this information to answer your questions in a comprehensive and informative way.

Summarization: The model can be used to summarize text. The summaries produced by the model will be concise and informative, and they will capture the main points of the original text.

Creative writing: The model can be used to generate creative text, such as poems, stories, and scripts. The text generated by the model will be original and engaging, and it can be used for a variety of purposes, such as entertainment, education, and marketing.

These are just a few examples of the results that can be obtained from a Large Language Model running on a desktop computer. As these models continue to develop, we can expect to see even more impressive and useful results in the future.

* * *

(Me)  What is the minimum Internet connection speed needed for the effective use of an LLM on a desktop computer?


(Bard)  The minimum internet connection speed needed for the effective use of an LLM on a desktop computer depends on the specific LLM and the tasks you want to use it for. However, a good rule of thumb is to have an internet speed of at least 25 Mbps. This will allow you to run most LLMs without any major problems. If you want to use an LLM for more demanding tasks, such as generating realistic text or translating languages, you may need an internet speed of 50 Mbps or more.

Here are some additional tips for getting the best performance from your LLM:

Use a wired internet connection instead of Wi-Fi. This will give you a more stable and consistent internet connection.

Close any other programs that you are not using. This will free up resources on your computer and improve the performance of your LLM.

Use a high-quality LLM. Some LLMs are more demanding than others, so it is important to choose one that is well-suited for your needs.

I hope this helps!

* * *

(Me)  Estimate the costs of running an LLM on a desktop system including internet connection services.


(Bard)  Sure, here are some estimates for the costs of running an LLM on a desktop system, including internet connection services:

Hardware: The cost of the hardware will depend on the specific LLM you want to run and the specifications of your desktop system. However, you can expect to spend at least $500 for a decent GPU and $1,000 for a high-end GPU. 

Electricity: The cost of electricity will depend on your local electricity rates and how much you use your desktop system. However, you can expect to pay an additional $10-20 per month for electricity if you use your desktop system for several hours per day.

Internet connection: The cost of your internet connection will depend on your ISP and the speed of your connection. However, you can expect to pay an additional $10-20 per month for a high-speed internet connection.


Total monthly cost:

Low-end: $70-100
Mid-range: $100-150
High-end: $150-200

It's important to note that these are just estimates, and the actual costs may vary depending on your specific circumstances. For example, if you live in an area with high electricity rates, your electricity costs will be higher. Similarly, if you use your desktop system for a lot of other tasks, your electricity costs will be higher.

Overall, the costs of running an LLM on a desktop system can be significant. However, if you need to use the LLM for a lot of tasks, it may be worth the investment.