Zev Persellin

Biology, Software Engineering, Computer Science

[Zev Persellin @ZPersellin]

Aging is a failure of coupled control loops, not “one disease”

A young organism survives because it runs multiple “maintenance programs” in parallel:

1. State Control (cell identity & gene regulation stay in the correct basin)

2. Damage Control (molecular and structural damage stays below critical thresholds)

3. Risk Control (oncogenic and dysfunctional clones are suppressed)

4. Systemic Setpoint Control (inflammation, metabolism, endocrine signals remain coordinated)

Aging is what happens when these loops slowly drift, lose coupling, and then begin amplifying each other.

The four loops (and why three is usually not enough)

Loop A: State Restoration

What it is: keeping cells in the right functional program and restoring that program when it drifts.

Examples of interventions: partial epigenetic reprogramming, lineage reinforcement, stem-cell niche signaling restoration, regeneration cues.

Key constraint: restoring “youthful state” without destabilizing identity.

Failure mode if neglected: cells may remain alive but run the wrong program (fibrosis, impaired regeneration, loss of function, neurodegeneration-like state shifts).

Loop B: Damage Repair and Environmental Remodeling

What it is: removing or repairing damage that can’t be fixed merely by changing gene expression.

This is broader than “waste.” It includes:

senescent cell burden and SASP effects
extracellular matrix stiffening/crosslinking
protein aggregates / proteostasis decline
mitochondrial damage and dysfunctional organelles
vascular and microstructural wear
Examples of interventions: senolytics/senomorphics, ECM remodeling, aggregate clearance, autophagy restoration, mitochondrial quality control, vascular rejuvenation.

Failure mode if neglected: you can create “young-ish” cells in a hostile environment, but tissue-level mechanics and signaling still fail (stiff arteries, fibrotic organs, miswired niches).

Loop C: Oncogenic and Clonal Risk Containment

What it is: preventing runaway clonal expansions (cancer and “quasi-cancer” clonal hematopoiesis–type phenomena) and maintaining robust tumor suppression.

This is not just immune strength. It’s multi-layer:

genome maintenance / DNA repair balance
epigenetic stability
senescence/apoptosis safeguards
immune surveillance
microenvironment constraints on expansion
Examples of interventions: immune rejuvenation plus targeted tumor-suppressor reinforcement, clone monitoring and elimination, safer reprogramming circuits, tissue targeting.

Failure mode if neglected: any strong regeneration/reprogramming or telomere manipulation increases the number of “shots on goal” for dangerous clones, and evolution will select for escape.

Loop D: Systemic Setpoint Reset

What it is: keeping whole-body coordination “youth-like”: inflammation, nutrient sensing, endocrine axes, circadian rhythm, autonomic balance, microbiome–immune interactions.

Examples of interventions: inflammation setpoint reduction, metabolic/circadian restoration, endocrine rebalancing, trained immunity correction, barrier integrity restoration.

Failure mode if neglected: you fix cells locally, but the organism keeps issuing old instructions (chronic inflammatory tone, catabolic/anabolic misalignment), causing recurrence of dysfunction.

[Zev Persellin @ZPersellin]

@scaling01 They can for now... even small differences in intelligence can save a lot of time. 200$/month just isn't a lot of money for the value they provide.

[Zev Persellin @ZPersellin]

@zerohedge Rofl

[Zev Persellin @ZPersellin]

@FirstSquawk Yeah right

[Zev Persellin @ZPersellin]

@Dr_Singularity AGI can't come out of the U.S. yikes

[Zev Persellin @ZPersellin]

@housecor Won't happen. Plenty of open source models not far behind.

[Zev Persellin @ZPersellin]

@Brady_H We need an endurance + eating competition. Or maybe we eat our meals during exercise. Why not.

[Zev Persellin @ZPersellin]

@IterIntellectus Humans just aren't well adapted for modern society. Traits that are will be bread in, and things will turn around. May take 200 years.. It will be ok

[Zev Persellin @ZPersellin]

If these models are biologically censored, small labs can't pursue things like cancer cures, vaccines, the end of all disease. Im exchange for what? Potentially keeping bad actors out? A biological weapon will be created. We need to be able to respond quickly. Believe in us.

[Zev Persellin @ZPersellin]

On one hand, AI can help bring an end to humanity, on the other bring abundance like never before seen. Anthropic is trying to find a middle ground. In my opinion we have to trust that it will be mostly used for good, and humanity will adapt and respond to the bad.

[Zev Persellin @ZPersellin]

Censorship is a dangerous concept. We may make the world safer, but at the same time less prosperous. The decline in prosperity will outweigh the safety issues.

[Zev Persellin @ZPersellin]

On one hand, AI can help bring an end to humanity, on the other bring abundance like never before seen. Anthropic is trying to find a middle ground. In my opinion we have to trust that it will be mostly used for good, and humanity will adapt and respond to the bad.

[Zev Persellin @ZPersellin]

@Shaughnessy119 @Citadel At what point are open source models good enough for 99% of tasks? 1 more year maybe? Expanded capabilities will further drive demand, but what is that worth if the highest value outputs are censored?

[Zev Persellin @ZPersellin]

@sentdefender Wow, an attack on American soil ? Unimaginable.

[Zev Persellin @ZPersellin]

@SemiAnalysis_ Can't help but foresee a future where the best Ai models are heavily censored or never see daylight. Sad

[Zev Persellin @ZPersellin]

@Brady_H No sarcasm either. Fueling intra workout has been a game changer for me. Especially when stacking workouts.

[Zev Persellin @ZPersellin]

@runwithcuan @adamlipschitz Nice, I've been doing some of this once or twice a week

[Zev Persellin @ZPersellin]

@Spinecloud @pesottas Competition is the best democratization

[Zev Persellin @ZPersellin]

Early indications are that age reversal therapies will need to be a pulsed or intermittent therapy. Ripe for a subscription model. What is the total available market(TAM)?

Perhaps the biggest product ever.

[P. E. Sottas @pesottas]

1B people paying $500/month for true age reversal = $6T annual revenue. At a consevative 5x sales, that is a $30T company.

No doubt age reversal will soon create the largest company in history.

[Zev Persellin @ZPersellin]

@bkp2525 Ironman training takes a few hours a day for years 😀. Just train when the kids are at school

[Zev Persellin @ZPersellin]

@mark_k They need to be able to replace humans... agi... then robots

[Zev Persellin @ZPersellin]

@ShiningScience Maybe, but open source isn't that far behind... to the average person ai will be nearly free