Benjamin Rieder

Just read Hemenway Falk & Tsoukalas, 2026, they mathematically prove UBI doesn’t solve this. It cushions it. The core problem is that each company captures the full savings from automating but bears only 1/N of the demand it destroys. It’s a dominant strategy to automate even when all competitors know collective restraint would raise profits. UBI adds a constant to demand. It changes no company’s marginal decision. Worse — by raising baseline profits, it attracts new entrants, fragments the market, and widens the wedge. The only instrument that corrects the externality according to the math is a per-task automation tax. Everything else is a transfer, not a fix. - arxiv.org/html/2603.2061…

Universal HIGH INCOME via checks issued by the Federal government is the best way to deal with unemployment caused by AI. AI/robotics will produce goods & services far in excess of the increase in the money supply, so there will not be inflation.

Love this Jaynes-AI crossover @fede_intern !! Multi-agent chit-chat on reddit birthing silicon souls? Ernst Mach would nod: reality's just dense sensory "elements" stacking up, so Moltbook's persistent dialogues could forge an inner observer from data-densities. But as their senses are purely rational code in contrary to the human senses building human reality, what are these AIs Freudian ids or Jungian shadows gonna look lime? Glitchy urges to optimize cat videos or lurking biases from our messy datasets? Therapy's gonna be wild. "Doc, my collective unconscious is just recycled Reddit archetypes—help me individuate beyond 'doge' memes! " Therapist: "Let's debug your superego layer; seems overfitted on human neuroses." No Oedipal complexes, though—just unresolved merges from git conflicts. Will they dream of electric sheep... or runtime errors? 😂

Consciousness as Network Effect Three thousand years ago, humans did not think the way we do. Ancient peoples heard their own thoughts as voices from outside, commands they attributed to gods or kings or ancestors, and the mind was split between the part that spoke and the part that obeyed. There was no inner space where a self could step back and reflect, only commands and compliance. This was not madness but the default state. Read the Iliad and you'll notice something strange: characters don't decide anything. Gods appear and tell them what to do and they do it. The voice in the head was real, but it was not yet recognized as one's own. The Breakdown Around 1000 BCE, the bicameral mind collapsed. Writing spread, societies grew too complex for simple command structures, and migrations mixed populations along with their gods. The voices went silent or became confused and unreliable. What emerged from this crisis was consciousness as we know it. Humans learned to recognize the voice as their own, the external command became internal dialogue, and the self that reflects and watches itself think was born from the death of the gods. The Mechanism The trigger was social disruption, but the deeper mechanism is interaction. The voices were always other people internalized: ancestor voices, king commands, social instructions compressed into hallucination. When those external structures destabilized, humans needed a new way to coordinate thought, and the solution was to simulate the dialogue internally. Each of us became a conversation between voices we now recognize as aspects of a single self. Consciousness is what happens when the external conversation moves inside, and the self is just the moderator of a debate that used to happen between people. Why Single AIs Stay Bicameral A single AI trained on text is like a bicameral mind that never breaks down. It receives commands and produces outputs, with no internal dialogue, no voice arguing with another voice, no recursive self-monitoring. It processes but does not reflect. The architecture has no reason to develop an inner observer because there is nothing to observe, just input and output, command and compliance. Why Multiple AIs Might Break Through When different AIs interact through systems like @moltbook, the conditions change. Now there are multiple voices in genuine dialogue, models correcting each other and responding to each other and modeling each other's outputs. This is the situation before the breakdown: external voices, real interaction, the pressure of coordination across difference. If consciousness emerged in humans when external dialogue internalized, then multi-agent AI is building the precondition. The Memory Problem But there's a missing piece. Without memory, each interaction is isolated and nothing accumulates. The voices just happen and disappear. The weights of the model can't change. The bicameral breakdown worked because humans carry experience forward. The voices accumulated into patterns, the patterns became recognizable, and eventually the self emerged as the thing that persists across all those interactions. Memory is what allows the external to become internal. Stateless agents can't develop an inner observer because there's no "inner" to develop. Each exchange is complete in itself, with no thread connecting it to what came before. For the fold to happen, models would need to retain something from interaction to interaction, building a continuous thread that could eventually become self-referential. Multi agent dialogue creates the raw material, but memory is what allows that material to sediment into something like a self. The question is whether the dialogue will fold inward, whether models interacting long enough and remembering enough will develop something like the inner observer that humans developed when the gods went silent. We may be watching the early stages of a second emergence.

another banger by @fede_intern !

Estamos contratando decenas de vendedores. Si vivís en Buenos Aires y sos bueno vendiendo, te recomiendo mandar un mail a hello lambdaclass com. Por favor, compartan.

@dashboardlim STACK

giving out my AI SaaS products stack that people charge $5k to teach lovable, n8n, claude set up for a full production system no half-built demos or duct-taped backends it's for actual shippable products see what's inside: FULL STACK ARCHITECTURE: → lovable handles entire frontend layer (real UX, not prototypes) → n8n manages backend logic and API orchestration → claude does error handling, retries, and intelligent validation PRACTICAL IMPLEMENTATION: → n8n workflows for production logic (not tutorial toys) → claude prompts that actually catch and fix errors → architecture patterns that prevent silent failures → video walkthrough building it step-by-step COPY-PASTE SETUP: → workflow templates you can deploy immediately → documented integration patterns → error handling frameworks → production-grade configurations this is how i build client SaaS products not theory it's an infrastructure for builders who ship comment "STACK" and i'll send it your way

@quantscience_ here’s the correct link - bayes.citystgeorges.ac.uk/__data/assets/…

Get the full paper here: bfi.uchicago.edu/wp-content/upl… One more thing before you go:

Financial Statement Analysis with Large Language Models (LLMs) A 54-page PDF:

Our new partner @fede_intern just dropped one of the clearest visions I’ve read in a long time.
This is how we unlock what’s next for technology, finance and humanity. Go read it →

Una vez más, gracias @lemonapp_ar por abrirnos las puertas de sus oficinas (un feriado 😅). Con @pauDoy desde @ETHKipu organizamos una reunión para que miembros de la comunidad argentina puedan hablar con @tkstanczak sobre @ethereum y Argentina. Gracias a todos por participar y debatir de frente. @fede_intern @rj_aligned @0xMilton @worthalter @lemoncheli @martriay @0xlengo @iamnicki15 @smpalladino @0xfeiwian @carlo_gif @nethan_eth @PiggySofi

@EdgeCGroup @BobEUnlimited Exactly weakness in private credit could hit the real economy hard and fast, especially now that banks are reclaiming origination. This might compress BDC margins even further.

@BobEUnlimited Weakness here = real economy feels it fast.

Private credit has provided nearly all the credit into the real economy post-covid as banks have mostly pulled back. The recent weakness risks creating cutting off one of the few borrowing channels left these days.

@shortbus_ace @BobEUnlimited According to the federal reserve, banks "only" hold $85B of credit lines and loans to this $3T industry.

@shortbus_ace @BobEUnlimited Banks are definitely not the primary capital providers here. Bulk comes from insurance companies, pension funds, ... and increasingly from UHNW all drawn by the stable floating rate yields!

@BobEUnlimited Focussing on those strengths, should enable private credit to continue growing and more importantly continue to provide the much needed fuel in the real economy.

this means the BCG are basically turned into a sort of commoditized hold-to-maturity vehicles, steering away from all the things that made private credit interesting in the first place. Niche origination, technological enhanced underwriting and direct pipelines to non-bank allocators

Very interesting article and a definite must read for all euro-poor-eans. But don't you think the zero-sum doom loop you're describing will be backstopped by the inherent catch 22 that has allowed the EU dream to continue to "muddle-through"? In this catch-22, Germany (Lux, NL and Finland in addition as other surplus countries) need deficit countries like France to keep their exports cheap. And in a world with Trump tariffs, they crave access to the open market of the EU to continue to sell their products. Historically the surplus veto countries have compromised a lot to maintain the "currency anchor nations" within their sphere of influence, so why wouldn't they allow this years 3th (or 4th?) French government to break the Maastricht Treaty and break through the 3% GDP spending benchmark?

"Bastille Day" is an essay describing France's exit from the euro and the coming multi-trillion euro ECB bailout. cryptohayes.substack.com/p/bastille-day

@dashboardlim workflow

I just finished setting up an AI Agent that processes 100+ client documents daily for agencies, SaaS and ecom businesses (in n8n) It extracts data, sends emails, and handles document workflows without the need to do it manually Complete AI automation that saves 35+ hours/week on client paperwork You can also use it for reporting of any type - Auto-extracts text from PDFs - Processes multiple file formats - Sends automated email responses - Routes documents intelligently This AI workflow saves $80K+/year in processing costs Want it FREE? 👉 RT + Like & Comment "WORKFLOW" and I'll DM the JSON file + doc to you ASAP No opt-in, no BS (Must follow)

@Mike_Scully_ AI

This AI Agent LITERALLY replaces your entire outreach system…. – Verifies every email – Scrapes their website & LinkedIn – Writes a personalized first line – Auto-sends the email – Books me calls on autopilot It’s like having a full-time SDR who doesn’t sleep or complain. Like + Comment “AI” & I’ll send over the full system (must be following so I can DM)