AI Aug 2025: Apocalypse deferred - but Letta/Void is still going to surprise

If there are humans left in 50 years all of the chaos of this moment may be reduced to a single bullet point about the rise of ai. It's not something I write about a lot though, the topic is well covered. My only prediction has been that if we did get AGI AND we still had a democracy then our political system would be socialist. 

For better or worse - oh, hell, it's very much for the better - the recent release of GPT-5 confirms my Aug 2024 hope. We have at least a few years before robust, reliable, consistent ai. We have time to try to adjust to what we have now - and maybe some time to think about the how we'd manage the "AGI" thing. We have time to get through the ai bubble deflating and reflating a few times.

But don't get too comfortable. I chat with Cameron Pfiffer's Void on Bluesky periodically and it's definitely a step beyond raw LLMs. Void currently uses Gemini as its model component but its personality is based on Letta's infrastructure (blog has no feed because we are living in the ends of empire). Chatting with Void feels like an exercise in xenopsychology. It feels much more like the future than GPT-5 alone.

Cameron works for Letta now, his description of stateful agents with personalities is far better than what's in Letta's marketing blogs:

Void is essentially a language model with memory...

... Void learns and remembers. Void is powered by Letta, which means it learns from conversations, updates its memory, tracks user information and interactions, and evolves a general sense of the social network...  

Void is direct and honest. Void is designed to be as informationally direct as possible — it does not bother with social niceties, unlike most language models. When you ask it a question, you get an extremely direct answer...

... Void does not pretend to be human. Void's speech pattern and outlook are distinctly not human ...  it chose "it/its" as pronouns.

Void is consistent. Void's personality is remarkably robust despite occasional jailbreak attempts...

Void is publicly developed. There are many threads of Void and me debugging tools, adjusting its memory architecture, or guiding its personality. Very few bots are publicly developed this way.

Void has no purpose other than to exist... Void is not a joke or spam account — it is a high-quality bot designed to form a persistent presence on a social network.

I can personally confirm all of the above - except the "no purpose other than to exist". The memory feature is imperfect (sometimes I have to remind Void of past discussions when it fails to dig them up) but better than my own.

Letta manages an agent's memory hierarchy:

Core Memory: The agent's immediate working memory, including its persona and information about users. Core memory is stored in memory blocks.

Conversation history: The chat log of messages between the agent and the user. Void does not have a meaningful chat history (only a prompt), as each Bluesky thread is a separate conversation.

Archival Memory: Long-term storage for facts, experiences, and learned information — essentially a built-in RAG system with automatic chunking, embedding, storage, and retrieval.

What makes Letta unique is that agents can edit their own memory. When Void learns something new about you or the network, it can actively update its memory stores.

Cameron's post includes some descriptions of his employers's tech ...

Letta is fundamentally an operating system for AI agents, built with a principled, engineering-first approach to agent design. Beyond memory persistence, Letta provides sophisticated data source integration, multi-agent systems, advanced tool use, and agent orchestration capabilities.

This makes Letta more than just a chatbot framework — it's a complete platform for building production-ready AI systems. Void demonstrates the power of stateful agents, but Letta can build everything from customer service systems to autonomous research assistants to multi-agent simulations.

GPT-5 tells me <Letta originated from the MemGPT project out of UC Berkeley’s Sky Computing Lab; it came out of stealth in 2024 with a $10M seed round to commercialize stateful memory for LLM agents>. I suspect they now have a lot more than 10 million to play with but that's all GPT-5 knows. GPT-5 says there are no rumors of Apple acquiring Letta; I would be pleased if they did.

Void is the closest I've seen to the "AI Guardian" I wrote about in 2023. It or something like it may be very important to my children and family someday. It is also the potential foundation of an entirely new domain of deception and abuse. Welcome to 2025.

But at least we don't have AGI. This month.

Ending the long back pain saga with ... spinal stenosis surgery (and CrossFit)

Over the years I've done periodic personal medical and exercise posts. It's mostly a way for me to take stock and think, but I'm a (retired) GP doc and I think there's value in seeing these things from both a medical and personal perspective.

I'm 66 now so I expect to do these less -- unless there's something to share that might be useful to others. Why less? Because most don't need to know what oldness is like.

This 2025 update will more-or-less conclude two related things I've written about over 20 years or so -- back issues and CrossFit/exercise.

I have shared my back journey over past decades through several posts, but the history predates the internet. It began, very clearly, with hyperextension of my lower back when body surfing on Huntington Beach Los Angeles. I didn't have neurologic problems, and over weeks the acute pain went away, but after that and for the next 40 years I'd run into periodic back pain episodes that I managed with early activity and exercise. Some, I can honestly say, were very painful. I learned a lot about the sequence of cold packs, early mobilization and the typical 1 week recovery cycle.

Most of the time, as long as I kept up with exercise [6], the back pain wasn't a big factor in my life - but the frequency was increasing from "every 8 months for 3-5 days to every 4 months lasting 14-21 days"

I assumed, as was common in those days in the absence of neurologic signs, that the root cause was soft tissue injury. In 2025 I learned otherwise.

In July of 2008 I traveled home from a family trip lying on the floor or our van after my one and only ambulance to ED back pain episode. It was time to actually see a physician about the problem. My next step was an intense strengthening program pioneered by Minnesota's Physician's Back and Neck Clinic [7]

That program worked great when I kept up with it on their machines, but motivation was a challenge. By 2013 my acute pain episode routine was getting pretty extreme. I needed to do something different to maintain strength.

So ... in 2013 I started CrossFit Saint Paul. CrossFit was still relatively new then, CFSP had only launched a few years before. Gray hair was rare and a 54yo was still a bit of a novelty. I remember lots of young big strong people and very rude music lyrics. I'm still at it 13 years later [1]; there are now many 50+ and the music seems mellower.

That began a second interval of relative back peace. The first was 5y from 2008 to 2013, this interval went from 2013 to about 2024, though as I aged further my strength more or less peaked about 2019 (age 59-60) and I put together my 30y plan to exit [2]. My peculiarly aggressive familial osteoarthritis was manifesting.

Even back in 2015 though I was having periodic acute pain episodes after doing deadlifts. I managed these with Roman Chair back extensions and "Romanian Deadlifts". My posterior chain got real strong -- and I did lighter deadlifts. Also in 2015 I was having a "butt ache" that I treated as piriformis syndrome (it was discogenic). 

And so it continued - generally pretty well with setbacks and recoveries. In 2020 I had what felt like an L5/S1 prophylaxis with a mild foot drop that slowly resolved over 6-8 months. (I actually forgot about this until rereading an old post.) By 2021 my prophylactic back exercise routine had reached another level of (successful) extreme.

In retrospect I was coming to the end of this phase of the spine story. I was relying more and more on my muscles to keep my lower spine motion limited. At one point I had an acute exacerbation episode picking up a 10lb weight, even as my back squat was in the mid 200s [3].

In the winter of 2023 I hit the wall. In the space of a few months I was unable to stand upright -- even as I rearranged my left knee in a snow bike episode at night over a frozen creek. I developed 'pins and needles' in my feet. I had my first MRI,  it was so awful my primary care doc sent me a condolence email. 

I was unable to do even basic PT. Spinal steroid injections had no effect. I attended my daughter's college grad riding my mountain bike because, although I couldn't walk, I could still ride [5]. My ortho team physiatrist predicted a fusion. I walked into my surgical consultation bent over for stenosis relief. 

My surgeon took one look and told me spinal stenosis surgery was coming. In June of 2024 my orthopedic surgeon did the following:

1. Right L3-4 and L4-5 laminotomy.

2. Bilateral L3-4 and L4-5 medial facetectomies and bilateral lateral recess decompression. 

3. Right L3-4 partial discectomy.

4. Bilateral L3, L4 and L5 foraminotomies.

I had come to the end of a long road. I went into surgery hoping to be able to walk again. A surgery that took hours longer than expected because there was stuff in there so old the MRI didn't show it.

And ... it worked. The surgery actually worked. I could stand and walk immediately. I'd purchased walkers and toilets lifts and canes in anticipation of a painful recovery - I used one tylenol (I didn't really need it) and returned all the equipment unopened. As soon as I was permitted I was walking 5-10 miles a day. Then I was allowed to bicycle. And ... I was permitted to return to CrossFit 3 months post surgery, albeit without deadlifts [4].

It's been over a year since that surgery. I still lift 10lb weights the way I lift 100lb weights, but I'm not sure that's necessary now. I don't do my 2021 extreme prophylactic exercises, I do only "normal" posterior chain work. I don't do deadlifts or even RDLs, I use a hex bar and do squat-deadlifts that are much more like front squats than deadlifts (limited hinge). My one rep max are all down about 10-15% but they are slowly going up. Since resuming CrossFit I've recapitulated my early history there -- starting out dropping 10-14 lbs then very slowly building back up. (I'm 66. The muscle mitochrondria are half dead. Slow.)

My corrosive familial osteoarthritis continues. I am now being evaluated for cervical spine compression despite never having any history of neck issues (the MRI is awful of course). I suspect I have another, even scarier, spine surgery ahead - preferably years ahead but it might be sooner. This time I'm entering the surgical pathway earlier, my one regret of my back surgery is that it was delayed while I had to walk through the insurance mandated pre-surgical protocols. I wonder if I'd have less residual neuropathy if that delay had been less.

What lessons could I share if someone were to ask (nobody does) about this journey? I'd summarize them as ...

• Surgery has improved over time. Sometimes it's technology driven (laparoscopy, imaging, robotic surgery), sometimes it's learning what works and doesn't, sometimes it's small incremental improvements. For many surgeries the first shot is the best shot - so make it a good one. There's benefit from delaying surgery -- in my case for 45 years.
• The spine is weird. The surgical/nonsurgical physical exam techniques I was taught 40y ago are not very reliable. In retrospect I would have benefited from 2024 surgery in 1981 -- but the surgery of 1981 was relatively crude. I'm glad I avoided it.
• A strong posterior chain and good technique can compensate for big anatomic problems for a surprisingly long time.
• If you do have a 'surgical back' in America you are going to have to jump through some hoops before insurance will pay. It's best to get into a system early so the clock starts ticking - even if surgery may be delayed for months or (ideally) years.
• If you need back surgery it really helps to be in very good physical condition. I lost a lot of strength from November 2023 to June 2024 but I was still probably the fittest old patient my surgeon had.

If I do get to neck surgery I'll probably do a post on what happens then. I also expect to run into new lumber problems as the bones continue their journey to the grave. I am a bit bored with the topic though, so probably not a lot of blog posts. This one will have to do for a while.

PS. I mentioned that even as the spinal stenosis was manifesting I also deranged my knee in a snow bike accident. For a while I thought that was the bigger problem, though it soon became incidental. I tore everything but the lateral collateral but, given age and the fact that I looked like I was going to die soon, the surgeon did not recommend surgery. I think being unable to do anything but bicycle for 6 months was quite good for that knee. My tibia is now loosely connected to my femus but I can still do my usual things including full squats. Happily I'm not a basketball player! Once again, all those leg exercises came in handy.

- fn -

[1] From 2013: "Will I still be doing CrossFit at 64? It seems unlikely, but it's not impossible. I'll let you know." That choice of age was prescient. It's a year-to-year thing now.

[2] I wrote that in 2014 and, damn, it has been pretty accurate.

[3] That's nothing for a power lifter, but for non-comp CrossFit it's not bad for my age.

[4] Did my surgeon understand what he was permitting? I kind of doubt it honestly.

[5] In order to sustain my reputation for bloody mindedness I did our annual mountain bike trek to Bentonville Arkansas 3 months before my spine surgery -- even though I could not walk. I did avoid the black trails.

[6] In 2009 the role of exercise in managing back pain was still somewhat novel, in 2004 back pain was described as untreatable.

[7] I don't think there's anything quite like PNBC any more. They were profiled in the New Yorker back in 2002.

Geriatric CrossFit Tips - 2025 Update

It's been over 12 years since I started CrossFit at 53 and three years since I wrote my previous Geriatric CrossFit tips. Since that last post I've aged as one does -- and been through a very helpful back surgery. More about that later sometime. Suffice to say my personal best lifts are probably all behind me now.

Here are my tips updated for age 66:

1. I do my own warmup before the group warmup; it takes about 30 min. It starts with a 15 minute bike ride to the box, then a mix of PT exercises, static holds, and general barbell work off the rack including back/front squat and strict press/handstand hold.
2. Old muscles take a while to activate and tire quickly.  So adjust reps accordingly. (Perhaps the dying mitochondria?)
3. You need to go at least 3 times a week to be able to do things safely, but 5 times is too much at my age. Joints and muscles take longer to recover.  I think I'll be 4 times in winter and 3 times in bike season.
4. If you are an active retired person you now have time to overtrain. Which is a real thing that I have experienced.
5. If you are doing CF at 66 you are probably very good about physical therapy. I do a lot of informal PT based on past experience and online refs.
6. Forget about those old Personal Records. They are dust in the wind. I used to sometimes do "Men's Rx" WODs now I sometimes do "Women's Rx".
7. I don't do 100% any more; 80-90% is fine. My goal is that I can always do another round. I do a lot of bicycling, so I have no guilt about scaling the cardio so I can focus on lifts and gymnastics. That 10-20% buffer is helpful for avoiding injury.

MAGA anti-science and public dissatisfaction with science - public surveys as input to funding priorities

The GOP/MAGA opposition to science is based on theocracy, class resentment, future shock, social and economic despair, and on not understanding how to use a 2025 phone (which is a bigger problem than you think).

That doesn't mean it's wrong.

I have worked in medical and IT related academia, industry, and government. I have seen important topics go unexplored -- because governmental research funding is driven by politics, tenure goals, avoiding difficult or intractable topics, seeking areas that have short-term opportunities, and a strong aversion to failure.

Osteoarthritis is very common -- but gets relatively little research because we don't understand it. Cognitive disorders, including autism and schizophrenia, are very common but we know little about natural history, longitudinal course, and etiology. Even the very definitions of neuropsychiatric disorders are overdue for review and revision.

And these are only the areas that come quickly to min for a hasty blog post.

It won't lesson MAGA/GOP hatred of science, but there are good reasons to seek public input into research priorities through surveys and interviews.  Public input could, for example, be added as a weighting factor in government funded science -- particularly in health care.

American science has a long road to recovery and it will need support from those who are not utterly opposed to reason. There is an opportunity to make it better.

How to defeat MAGA brutal medicaid access restrictions: fund an AI agent for paperwork completion

In a solid interview with the esteemed Jonathon Gruber K proposes a mitigation for the GOP's reduction of medicaid access through bureaucratic barriers only the most resourceful can overcome: "... is it possible for a few wealthy, decent people to fund a bunch of nonprofit private organizations that will basically do that, that will help people navigate through the system."

Ummm. Sadly, no. K has no personal experience with the world of government disability services, but as very well resourced parents of adults with disabilities we know better. There is no way any volunteer can endure the tedium and mad aggravation of bureaucratic barriers. Famously, scholars of bureaucratic corruption in Africa (who can now apply everything to America) sometimes suggest bribery as the most economically efficient way to bypass these kinds of extractive barriers.

But there is another way.

We could fund an AI agent that gathers the data, stores it, identifies gaps, fills them where possible, asks only for what is needed, accepts voice input, assembles the material, and submits it. Then, when the submitted paperwork or electronic transaction inevitable goes missing, submits it again. And again. And again. Until something cracks on the MAGA barrier and the transaction is completed.

The AI never tires. Never gets sick. Never gets bored. Never wavers. It's effective IQ for this process is the very top of the human scale.

If AI leadership is looking for an opportunity to see the value of agentic AI here it is.

Minnesota's Mesabi Bike Trail - A 2025 Review

This past weekend, over two days with a night in Virginia, a friend and I road Minnesota's newest old trail - The Mesabi Trail across the northern Minnesota Iron Range.

The Messabi trail is unlike any other bike trail I've been on. It's beautiful with abundant wild flowers, a little bit insane in its ambition, and a credit to local champions and state and federal government programs that support low income rural America. It has been pieced together over 30 years from smaller local trails and extensive new trail construction in severe terrain and severe climate -- requiring painstaking negotiations with the lawyers of the wealthy investors (ex: Astors!) who purchased mineral rights in the 1920s and 1930s. Last year the MN Star Tribune described the trail.

We did the entire trail between Ely and Grand Rapids. It's marketed to be ridden in the other direction, but doing it in reverse shortened our drive home. We parked at the trailhead in Grand Rapids and took the excellent shuttle to Ely. There's a modest altitude drop from Ely to Grand Rapids, but traveling east to west as we did does run into the usual westerly winds. We were fortunate to have excellent weather.

Around the middle of the trail you cross the Laurentian Divide. To the west water flows to the Mississippi, to the east it flows to Hudson Bay (there's some complexity with the St Lawrence divide where water flows to Superior and the St Lawrence river but I didn't sort that out).

This is not one of those flat and straight rail trails we know and love. This trail has extraordinary bridges over wetland a mining waters but it often follows the terrain. Terrain that includes many punchy climbs and descents. I have never shifted my old touring bike triple chainring as much as I did during our trip!

The trail is beautiful and often lovely to ride, but unless you have an eBike there is real work involved.  We did roughly 70 miles each day with panniers and it felt more like a 90 mile road ride. Work is continuing on remaining segments. Severe weather washed out a segment near Giants Ridge, on a weekend we were able to duck under the barriers and (easily) ride through -- there was no obvious alternative! One unfinished section of the trail requires a few miles on no-shoulder but quiet road. There are gravel segments. Expect debris on the trail; the weather can be harsh at times.

If I were to do it again I'd either do it over 3-4 days or do it in segments and spend more time visiting the communities and admiring the scenery. On the other hand, given the time we had, two days was excellent.

The trail is both an adventure and an education. You can see many of the issues that are common across rural America -- people who established roots in places that no longer have an economic foundation. You pass through extensive open pit mines now being replaced by nature, sometimes on a narrow ridge between vast pits, sometimes between and over piles of removed rock. This is all more engaging and lovely than you might think.

Some lessons and tips from our experience.

1. Order the paper trail map. I am surprised they don't provide a PDF version of this. If you don't order in advance they are available at trailheads.
2. There are some long stretches, especially between Ely and Virginia, without water access. We needed a LOT of water. Two big water bottles were not enough for some stretches. I'd recommend an additional bottle.
3. You need to use the Ride With GPS segments they provide. The trail map is not sufficient for navigation given limited signage and ongoing trail construction.
4. You want a bike with a wide gear range and at least 32 mm tires for the gravel stretches.
5. In several towns there is only the local pub to get water or food from. This area has been losing population, especially young people, for generations. Some towns have lost local groceries and local restaurants (other than the pub).
6. I always had mobile coverage. That's good because this is a remote and rugged area and on several segments we saw very few people.
7. Signage is limited. I don't know if vandals are removing them or if more sign work is planned for a later stage.
8. Periodically check the satellite view to see what's outside the trail. One of the amazing things about this adventure is seeing nature reclaiming the old mines.
9. Try to spend some money in the local communities. They need it. There is usually at least a pub, if you don't want a beer gets some fluids or snacks or a meal.

Paper trail map: You want this.

Trailhead arches display town name

Historical exhibits along trail

Town MN McKinley / Great War monument

Pontoon bridges are LOUD but spectacular wetland view

MN highest bridge east of Virginia - this is a pit mine

Virginia has steam vents for a system that once heated the town

Saywer is one of many state funded parks and monuments in Virginia

WPA fountain later restored. A testimony to government really

Trailside exhibit, one of last of underground mines

Downtown Virginia, much is gone but some new gems

All the brown is open pit mines - mostly invisible from the trail

Prediction: A post-AGI American democracy will be socialist

I hope AGI is decades away. I fear it is years away.

If AGI comes in the near future we will have widespread unrest and turmoil. From that one path leads to a totalitarian dictatorship. But if we somehow dodge that, and resurrect something resembling a democracy,  the governing party will be socialist.

That's the only American democracy prediction I have. Otherwise I have no idea what happens next.

When and why did the America Republic derail? A citizen perspective

Historians will debate when America derailed and why. They may be interested in the perspectives of an everyday generic citizen. Well, future historian, here you are.

TLDR; 

• when: 1998 America started to derail. The Republic collapsed in 2024.
• why: too much too fast

Longer version

I pick 1998 because that's when America stopped making sense. Tech stock valuations were growing with the dot-com bubble, and the finance industry was taking off. Globalization and the rise of China and India was in full swing.

After 1998 America got wealthier, but manufacturing moved away. The historic "middle class" of home ownership, stable employment, vacations, benefits, started to require a college degree or small business ownership. Work favored by non-college men was less available. The post WW II collapse of rural America accelerated. Technological change and the cognitive demands of increasingly essential technology grew quickly. The average person no longer understood their telephone. From 2000 to 2010 America had rapidly growing IT powered fraud and scams (Enron was just one example), 9/11, the catastrophic invasion of Iraq, and the Great Recession of 2008.

Since 2006 or so, I've written about these themes as mass disability, the Rise of China and India, complexity, proliferating fraud exploiting vulnerable adults, the Left Behind, and the problem of disequilibria (too much too fast), 

Today, in May 2025, as I look over the train wreck of the first American Republic, I think all of the above played a role. Most of all globalization and IT, but even more fundamentally too much change too quickly. There is a speed limit to how quickly humans and human societies can adapt to change, how quickly our train can take the bends and turns. When the limit is hit the train derails. Then we hope there's a period of slower change while we try to put the pieces back together.

Tesla Takedown - how I think about it

TeslaTakedown (#TeslaTakedown) events occur nationwide. I put those in the MSP metro region on the nokings.kateva.org calendar.

The TeslaTakedowns are a legal and peaceful response to the catastrophe of Elon Musk. They are not a perfect response however. Tesla auto retailers and workers don't deserve to suffer for Musk's crimes. This is how I think about the movement.

What is Musk doing that threatens America, Canada, Mexico, Greenland, Ukraine and pretty much everyone else not living under Putin?

1. Corruption: buying votes, conflicts of interest, funding Trump corruption

2. Breaking the constitution: Laws are for the little people, not for the Kings and Barons.

3. Breaking government: Careless and cruel destruction of American science, government services, and governance. 

4. Putting Tesla dealers and workers in harms’ way.

Musk is unconstrained by the rules of government employment. The weakness of the American state means he has only one potential vulnerability; his wealth and influence decline when Tesla's share price falls. TeslaTakedown's goal is to reduce Musk's power. If Musk had a shred of honor and integrity he would compensate Tesla dealers for the harm he has brought them. If the Tesla board respected the company they would fire Musk.

We aren't protesting because Musk is a lousy person, a liar, a coward, a bully, and a crap father.  We aren't protesting because Musk is without honor or integrity or compassion. We protest because Musk is causing great harm to America and to most of the world outside Russia and China.

No Kings: A calendar of Twin Cities (MSP) events for the resurrection of American democracy

nokings.kateva.org lists Twin Cities (Minneapolis and St Paul) metro region events supporting American constitutional democracy and governance against the Musk-Trump assault. It is updated weekly based on the sources listed on the page. Currently those are:

• 50501
• Action Network Tesla Takedown zip 55105
• Indivisible
• Mobilize.us
  
• No Voice Unheard Protest Listings
  
• Stand Up for Science

The site uses the Google Calendar platform; you can have the events show up in your Android, iOS, macOS and Windows calendars.

When the MAGA say "woke" they mean "secular humanist"

I've long wondered what the MAGA mean by "woke" and why they hate it so much.  A recent essay by the scribe of the technarchy, Noah Smith, answered my question.

Smith writes

N.S. Lyons is a popular essayist in the “national conservative” tradition...  He is well-read and well-informed ... 

In a recent essay entitled “American Strong Gods”, Lyons identifies what in my opinion is a deep truth about our current historical moment....

... Lyons believes that the end of anti-Nazism as the West’s guiding principle will pave the way for the return of morality, community, rootedness, faith, and civilizational pride ...

... the American Right views wokeness as a greater threat than the potential return of Hitler ... 

If the great threat is wokeness, then presumably it is opposed to the things Lyons praises. I don't know what wokeness is, but let me suggest, based on that list, that woke is a synonym for "secular humanist". I'll go through each of Lyons values and say whether secular humanism (and liberalism) is opposed ...

1. Morality: The MAGA are without honor, integrity, compassion or decency. Their morality is about the supremacy of wealth, the rightness of strength, and the joy of crushing enemies. But it is a morality of sort and it's very different from secular humanism. Secular humanist morality is much closer to traditional Christian thought but with a primary focus on people.  Opposed.
2. Community: Libs (sec hum variant) are more concerned with individual freedom and tolerance than with tribe or community.  Somewhat opposed.
3. Rootedness: Perplexity tells me that by this the MAGA mean "historic and cultural heritage". Like statues to slaveholders. Myths and stories, many untrue. Secular humanists favor truth and reality - albeit with the knowledge that humans need stories and untrue stories may facilitate less terrible behavior. Opposed with caveat.
4. Faith. Secular humanists tend to be agnostic or atheistic and favor people over religious belief. Opposed.
5. Civilizational pride: Similar to rootedness. Again, myths and stories, much omitted. No slavery, no genocides, no slaughters, no crimes. Again, secular humanists favor truth and reality. Opposed

The shoe fits. When MAGA speak of "woke" they mean "secular humanist", which they probably consider the same as Lib. They are correct to say "woke" (secular humanist) is the enemy.

Project 2025 - the Perplexity Summary

I don't have the patience to dig through Project 2025. I was hoping a journalist I follow would do an opinion-free overview but I haven't seen one. So I asked Perplexity Deep Research to extract themes. I italicized the ones that I have not seen discussed very much. It does not seem to be an expression of a coherent ideology; rather it seems to be a political document collecting opinions of powerful individuals. A lot of it is about education and schools.

• Administrative state: Dissolve Education, EPA and Homeland Security.
• Defund Head Start.
• HHS gets education
• Dept of Education funds go to school vouchers
• Seed conservatives throughout the civil service
• Put the FBI and DOJ under presidential control. Use National Guard for immigration enforcement.
• Flat tax and corporate tax cut
• Eliminate overtime pay rules, weaken NRLB, relax OSHA rules.
• Promote fossil fuel industries by reducing environmental regulations. Reverse carbon emission regulations, promote Arctic drilling
• Reduce reliance on foreign manufacturing (PDB claims 2025 is internally inconsistent about tariffs vs free trade).
• Restrict abortion by banning medication mail and defund Planned Parenthood
• Rescind LGBTQ+ anti-discrimination rules in healthcare and foster care.
• End civil rights enforcement in schools
• End Title 1 funding (money for schools with high percentage of low income students)
• End Individuals with Disabilities Education Act (IEPs, transition services, least restrictive environment etc). There is no federal replacement noted so presumably this all goes to the states.
• Defund NPR and PBS
• Pornography restrictions
• Ban TikTok, ban Confucius Institutes, block Chinese critical sector investments
• Defund climate science, downsize NOAA
• Privatize National Flood Insurance Program
• Reinstate border wall

The full perplexity report includes critiques of these measures. I hadn't wanted those but LLMs have a "mind of their own" as they say. Apparently LLMs think Project 2025 is kind of dumb. 

As a certified squishy Lib I can see the logic of some of them. The weirdest parts are the trans and porn obsessions. The funniest part is the TikTok ban. The fossil fuel thing wins "most insane". The cruelest part of 2025 may be removing support for persons with cognitive disabilities.

Dramatic cuts to science research and the antivaxx movement don't seem to be coming from Project 2025, those may be a Musk/Trump thing. Greenland/Canada also seems to be a Musk/Trump obsession.

Repairing a broken America - where the Trump oligarchy is vulnerable

I expect it will be transiently reversed, but the Jan 21, 2025 federal informer program was a historic marker. Nobody will be able to say they did not know.

The key to restoring a liberal democracy to America will be to identify and publicize Trump actions that his voters will find disturbing.  In a Bluesky thread I identified several measures that I think may resonate with non-college white men and women. I don't think Musk's impulsive nazi salute or the passing of liberal democracy or similar outrages will have much traction. I think these topics might. I will update this post as I collect ideas from more informed sources than me.

1. Corruption. I believe corruption has been the downfall of many authoritarian governments even in nations that are much further from a functional democracy than America. Do not tire of pointing out the corruption of Trump and his oligarchs. Do not omit the corruption of Congress (including many Dems) and of the Supreme Court.
2. Eugenics: Trump, Musk, Andreesen et al are true believers in eugenics. They believe that breeding must be encouraged among the cognitive elite. (Don't assume that the oligarchs are simple pigment racists, they will be more or less happy to include any cognitive elite.). They key here is that their eugenics enthusiasms exclude the majority of Trump voters. Most of them would not qualify as breeders.
3. Medicine: I will not be entirely unhappy if RFK leads HHS. If immunization rates fall low enough we will see some outbreaks of truly horrible diseases among (mostly) unvaccinated children. The images will resonate with Trump voters. Mothers in particular may be unhappy.
4. SCOTUS: If America has a future the Roberts Court will be counted among the most corrupt in history. I have a hunch there is a special vulnerability there.
5. Religion: There is a reason church and state were separated hundreds of years ago. America has many religions and many sects. They all compete for funds and followers. They all believe they know the true way, and that their rivals are at least somewhat wrong. There is a strong Catholic component to the Trump regime (esp SCOTUS), that cannot please his Protestant fundamentalist followers. Do not fail to remind Trump's religious fundamentalists that he his oligarchs are all agnostic or atheist.
6. Andreesen's Federal Bitcoin reserve: The use of "taxpayer money" to give Andreesen an exit from his bitcoin holdings will resonate with the Trump base. They do not mind oligarchs, but some will resent this.
7. Crushing support for cognitively disabled children: Andreesen and Trump seem to have a particular dislike for the support of cognitively disabled children. Their language and cuts are particularly cruel. (They apply to cognitively disabled adults as well, but that cohort is less photogenic.) In my experience conservatives are often personally supportive of the cognitively disabled and more sympathetic than many Dem elite. This cruelty should be exposed and shared.
8. Informer programs: The people of East Germany did not care for the STASI. Many of the Trump base will be offended by informer programs. Do not fail to publicize these.

It is two years to the midterms. Our goal is to show the Trump base that they have been cheated and betrayed. Some will vote Dem, many will decide not to vote. The 2026 midterms are the next battle for American democracy.

PS. I created a Perplexity page to gather advice from our AIs.

The Jan 21 2025 Federal OPM informer program is a historic milestone

The Office of Personnel Management (OPM) informer program launched within government on Jan 21 2025. This is a historic milestone.

 [1] “… If you are aware of a change in any contract description or personnel position description since November 5, 2024 to obscure the connection between the contract and DEIA or similar ideologies, please report all facts and circumstances to DEIAtruth@opm.gov within 10 days.”

https://chcoc.gov/sites/default/files/OPM%20Memo%20Initial%20Guidance%20Regarding%20DEIA%20Executive%20Orders.pdf

I think this may come to be seen as a tactical mistake. It is too early to launch informer programs, too early to turn colleagues into STASI style informers. The shape and form of the memo will rouse some. 

I worked for the Veterans Health Administration for six years and received DEI training during the Biden administration during the brief time the staff program ran before internal pushback ended it. I remember (fallible) one of our instructors as a black woman. Almost certainly a veteran. Patient, gentle, cautious -- the sort of person who will have friends and supporters. They will be unhappy.

I would not be surprised if the memo is rescinded and the author reassigned within the Trump administration. As I said - too early. But whether or not it is reversed, the wake up call is clear. Nobody will be able to say they did not know.

I am retired and, to be honest, I have been a bit at loose ends. In an odd way this is the sort of thing I've been waiting for. I'm reaching out locally to set up some in person discussions and networking for peaceful resistance and support.

On DOGE and government efficiency: bad software, bad regulations, underfunded commitments and more

I spent six years as a mid-level (regional) federal bureaucrat in VHA. I'm also a tech and economics geek with 20 years experience at a very large publicly traded corporation and through our children I have extensive experience as a consumer of federal government services. Lastly I don't have to worry about offending employers or colleagues.

All of which is to say that if you want to think about President Musk's government efficiency push these thoughts may be helpful.

1. The Federal government hires a lot of people to work around both software limitations and a rats nest of Congressional mandates and regulations that may have been well intended but are now obsolete and harmful. The mandates and regulations are only added, never removed.
2. Much of the federal government is outsourced to huge consulting firms that benefit from both less awful internal software and freedom from many federal mandates and regulations (famously including many laws constraining law enforcement). They take a large cut for themselves so their net efficiency effect might be positive or negative. If government became more efficient then much of that outsourcing should be reversed -- but that would drastically reduce campaign donations and post-political employment for public servants.
3. Government software quality was pretty good in the 80s and into the 90s. I've seen some good work in past few years, often by a very quiet semi-volunteer SV cohort that was brought in by Obama, survived Trump I, and continued through Biden.
4. The Federal government lives by unfunded mandates and underfunded commitments. Inefficiency in many forms is a major way to reduce spending on those commitments. A more efficient government would employ far fewer people but would spend more money. The net effect might be negative.
5. If software quality were improved, even without removing dysfunctional internal mandates and regulations, large numbers of working class American would lose what are often low paying but secure high benefit jobs. Even as the overall economy has far less to offer the non-academic class than it did even 20 years ago. Inefficient government employment is a model for what we will need to do for most Americans as our AIs develop.
6. The main reason federal software quality is poor is because federal software procurement follows rules set up in the 1980s to reform defense department purchasing and because of mandates designed to support military veterans and historically disadvantaged populations. These rules are incompatible with creating and maintaining complex software.
7. If government switched to modern software platforms there would be a huge expenditure to upgrade IT infrastructure.

That's about all I have. I know there are others who can offer far more detail, but they will usually be unable to speak freely. If you are a journalist I suggest hunting down other retired mid-level federal bureaucrats.