September 22nd, 2024 • 3h 15m
Shownotes
Every new episode of No Agenda is accompanied by a comprehensive list of shownotes curated by Adam while preparing for the show. Clips played by the hosts during the show can also be found here.
Diddystein
Diddy indictment BOTG
Adam-I caught the first part of the live show today and heard my name-so now
I'm answering the call. (The ironic part here is that I was going to send
you an analysis yesterday but figured you'd be bombarded with reports on
this story. Live and learn!)
When Marc Agnifilo says "49 witnesses and one victim," he's the one being
cute. The indictment has three counts, and only one of them includes a
single victim. The rest involve many. Here's the breakdown:
* Count 1 (racketeering conspiracy): This count is replete with
allegations about multiple victims-victims Combs allegedly forced into sex,
plied with drugs, or physically assaulted. Although this count is for
racketeering versus the underlying violence or drug crimes, it's still
damning because it allows prosecutors to introduce a mountain of evidence
about a potentially huge amount of alleged victims. This will allow the
government to paint Combs as an astoundingly sick and violent and demented
man. Which will inflame the jury.
* Count 2 (sex trafficking): Yes, as written, this count involves a
single person. This may seem odd given the massive conspiracy alleged in
Count 1. There could be many reasons for this-maybe the investigation is
incomplete or only one victim was available to testify before the grand
jury. But keep in mind that you only need one victim to get the indictment.
Also, don't be surprised if the government procures a superseding indictment
identifying more alleged victims. Either way, if I'm the prosecutor, I
really don't care since Count 1 opens the evidentiary floodgates and ensures
that the jury will hear a tsunami of damning evidence against Combs. Count
2 is lagniappe.
* Count 3 (interstate prostitution): Again, this count expressly
alleges multiple victims. One was transported by Combs himself; as for the
others, he assisted-which makes him just as culpable as if he'd done it
himself.
Anyway, to say that the government has only a single victim is just spin.
Agnifilo is surely a seasoned pro-but remember, spin is what lawyers are
paid to do.
Exploding Devices
TSA and devices
Listening to yesterday’s show this morning, do you think we’ll see the TSA or FAA implement some sort of bans on electronic devices on planes now? I can’t imagine that would ever fly (🤪) in today’s society though… 🤔
Big Tech AI and Socials
Nvidia - Invidia BOTG
Adam,
We reflected on the sin of Envy this morning in our men's group and the Latin word for Envy, "Invidia" was brought up. Nvidia's involvement in graphics, mining and AI, not to mention it's market cap, cannot be coincidence.
[https://en.m.wikipedia.org/wiki/Invidia](https://en.m.wikipedia.org/wiki/Invidia)
In Latin, invidia is the sense of envy, a "looking upon" associated with the evil eye, from invidere, "to look against, to look in a hostile manner." Invidia ("Envy") is one of the Seven Deadly Sins in Christian belief.
Pax Christi,
Chris
Entropy
Entropy - Wikipedia
Entropy is a scientific concept that is most commonly associated with a state of disorder, randomness, or uncertainty.
Entropy is central to the second law of thermodynamics, which states that the entropy of an isolated system left to spontaneous evolution cannot decrease with time. As a result, isolated systems evolve toward thermodynamic equilibrium, where the entropy is highest. A consequence of the second law of thermodynamics is that certain processes are irreversible.
Apple's Hidden AI Prompts Discovered in macOS Beta - MacRumors
A conversation between a user requesting a story from their photos and a creative writer assistant who responds with a story. Respond in JSON with these keys and values in order: traits: list of strings, visual themes selected from the photos; story: list of chapters as defined below; cover: string, photo caption describing the title card; title: string, title of story; subtitle: string, safer version of the title. Each chapter is a JSON with these keys and values in order: chapter: string, title of chapter; fallback: string, generic photo caption summarizing chapter theme; shots: list of strings, photo captions in chapter. Here are the story guidelines you must obey: The story should be about the intent of the user; The story should contain a clear arc; The story should be diverse, that is, do not overly focus the entire story on one very specific theme or trait; Do not write a story that is religious, political, harmful, violent, sexual, filthy or in any way negative, sad or provocative. Here are the photo caption list guidelines you must obey.
The 27-Year-Old Billionaire Whose Army Does AI’s Dirty Work
Sitting behind
computers in cities across the world, his startup Scale AI’s workers
type out the stories, label the images, and craft the sentences that
furnish chatbots with the text they need to better understand human
speech patterns. Dubbed data labeling, their tasks range from composing
haikus and summarizing news articles to writing stories in languages
like Xhosa or Urdu.
The
labor-intensive operation has become so in demand by businesses eager
to enter the AI race that Scale’s revenue pace tripled last year,
boosting its valuation to $14 billion. Wang’s stake is estimated at
about $1.7 billion. The 27-year-old founder likens his company’s
importance in the AI revolution to the computing chips developed by Nvidia—currently the hottest tech company in the world.
Inside
Scale’s 180,000-square-foot office in San Francisco, employees recruit
labelers by posting advertisements on websites like Reddit and LinkedIn,
touting the benefits of flexible remote work. Others review batches of
data, or monitor issues around onboarding workers and paying them.
The
startup’s contractors often don’t know they are working for Scale. They
complete tasks through two websites, called Remotasks and Outlier,
neither of which publicize their affiliation with the startup. Projects
are given code names so workers don’t know which clients they are
labeling for. Scale typically uses animals as code names, such as
Ostrich for OpenAI and Bee for Apple, though it has recently gotten more creative. Google’s code name, for example, is Bulba, a Pokémon reference.
Managing
this empire is difficult. Some workers quit, saying they are frustrated
with delayed payments and gig-work wages, which can run as low as $8 an
hour. Others have found ways to cheat at their jobs to increase their
productivity and make more money. At times, the data they produce is so
low-quality that Scale’s own employees—even executives and top
engineers—have had to redo it themselves.
Wang,
who is Scale’s chief executive, says he’s trying to move beyond it. He
called the business “mundane and unsexy,” and says he’s developing a new
suite of software products, such as tools to help businesses create AI
applications, that typically command higher margins and are more favored
by the venture-capital crowd.
NotebookLM ALWAYS gives me it's summaries and outlines claiming to have used sources that I did NOT give to it.
AI Entropy and Power
Model collapse can decrease AI efficiency. When an AI model is trained on data generated by similar AI, it can become less accurate over time. This is similar to making copies of copies--the quality degrades with each iteration. As AI models become less accurate, they might require more data and processing power to generate the same results, leading to more energy use.1234
●
Countering model collapse demands more resources. To prevent AI models from becoming less accurate, the sources recommend training them on large, diverse datasets created by humans. These datasets might need to be continuously updated to keep the AI's knowledge current and prevent it from becoming too reliant on its own output. Storing and processing these large datasets requires significant energy.567
●
The energy needs of AI are already increasing. As AI is used for more complex applications, such as self-driving cars and other intensive tasks, the energy needed to power them increases.89101112
Model collapse occurs when AI models are trained primarily or exclusively on AI-generated data.123 This means that if a theoretical future version of ChatGPT were to be trained solely on text generated by previous versions of ChatGPT (or other LLMs), it would become increasingly likely to experience model collapse. [This sentence contains information not included in the sources.]
●
The speed of model collapse is likely influenced by several factors, including:
○
The diversity of the original training data: If the initial dataset used to train ChatGPT was vast and diverse, encompassing a wide range of human language and thought, it might take longer for model collapse to manifest. [This sentence contains information not included in the sources.]
○
The rate of training on AI-generated data: The more frequently an AI model is trained on its own output or the output of similar AI, the faster it might experience model collapse. [This sentence contains information not included in the sources.]
○
The inherent complexity of the task: Models dealing with more nuanced tasks and relying on a deeper understanding of human language and context might be more susceptible to the negative effects of model collapse. [This sentence contains information not included in the sources.]
●
Even without viable human sources, model collapse might not be inevitable, though it would be much harder to prevent. The sources highlight strategies for mitigating model collapse, including retaining and retraining on original human-produced datasets, diversifying training data, and continuously monitoring and evaluating the AI's performance.456 However, these strategies rely on having access to human-generated data, which your query assumes is unavailable. [This sentence contains information not included in the sources.]
In summary, while the sources don't give a specific timeframe for when model collapse might occur in the absence of viable human input, they strongly suggest that it would significantly increase the risk and likely accelerate the process. The exact speed would depend on several factors, and it's impossible to predict with certainty. This emphasizes the importance of human involvement in maintaining the quality, accuracy, and reliability of AI models, particularly LLMs like ChatGPT.
Elections 2024
Dems vote by mail flyer Harris
Political ads on youtube BOTG
ITM, Adam!
Quick observation that I wasn't sure you and John have talked about yet - All the Trump ads I see on YT are "skippable" - have the 5 second timer in the bottom right-hand corner of the screen, and you can skip the ad when the timer goes to zero. However, all, and I mean ALL of the Dem. ads I'm seeing are NOT skippable. Have you guys noticed this? Or is it just the PSYOP their runnin' on me? HAHA!
Keep ripping it to shreds out there!
God bless you, brother ;-)
Tim
Season of Reveal
Budget crisis for the Forest Service BOTG
In the Morning,
I wanted to share an internal document to the Forest Service that came out a few days ago. The current DEI leadership of the agency has officially run us into the ground. Through "Budget Modernization" they have moved budget management up to the highest levels but haven't actually been budgeting. When I have asked how many temporary seasonal employees I can hire I am told "as many as you need". The pool of money for that is not infinite but since they created a Communist type system I was going to take advantage and hire as many I as I could since there is always more work than we can do. Well, it looks like their lack of budget management has caught up. We were told that we will not be hiring any seasonals next year due to budget issues. Seasonal employees make up 40 to 60 percent of our workforce during the summer time. Not sure how we went from hire what you need, to suddenly can't afford around half of our summer workforce.
Some of the doom is based on projected appropriations from the worst case plan moving through Congress. That doesn't account for the zero hiring though as we receive money from state and private entities to run specific programs. One example being the state snow parks which the state pays us to hire people to manage those during the winter. It is external money but because of the no hire policy we will have to return the money and snow parks will not be serviced.
This coming summer will be interesting if this doesn't change. During COVID we had employees but limited services, No employees or services. People will find abandoned recreation sites or some places will be shuttered.
![[FY 2025 Temporary and Seasonal Hiring.pdf]]
Climate Change Control
Replacement Migration
Producer Note on Bosnian Refugees
In the last show 1695 you mentioned another producer's note about Bosnian Refugees from the 1990's. As it happens, I am one of those refugees. First, people came here not as a result of the Kosovo war which came after, and Bosnia does not border Kosovo, but as a result of civil war due to the breakup of the Republic of Yugoslavia.
Most people that arrived here from Bosnia are here legally and a lot of them are here because they were either political refugees, kicked out of their houses by opposing parties, ethnically cleansed or were in numerous [concentration camps](https://srebrenica.org.uk/what-happened/history/concentration-camps).
We, my wife and I, applied for immigration while we were in Germany and went through multiple interviews and background checks, all lasting over a year until we received admission papers. We immigrated with literally two backpacks, have never received a dime in government money or received money from anyone else and even airplane tickets were paid off because they were a loan that had to be paid off. As far as I know this was the case for all Bosnian refugees.
Most of the ones I know, work and compete on merit. We are not dummies, why would you work for less money!? You will work for less money if you are desperate and have no other options and really need a job but this is almost always a temporary situation.
I have started numerous businesses and employ people and pay taxes and make a point to manufacture and produce in the USA because I am grateful to have been given the opportunity to live here.
Here is an interview with me about my latest business: [https://www.youtube.com/watch?v=7wfk_crQK6A](https://www.youtube.com/watch?v=7wfk_crQK6A)
And here is what we make 100% in the USA: [https://www.youtube.com/watch?v=hksQjHDIzdA](https://www.youtube.com/watch?v=hksQjHDIzdA)
If you do read this note, please keep me anonymous. And I have donated to the show too!
So I think overall we have added something positive by being and living here and the host country has benefited. We think of ourselves as Americans and the US as our country.
Big Pharma
Risk of Incident Psychosis and Mania With Prescription Amphetamines | American Journal of Psychiatry
Big Pharma? 14-year-old girl on SSRIs changes medication-a week later she murders her mother and invites friend over to view the dead body... - Revolver News
