Marty Cornish

CEO @workpackai

[Naval @naval]

The new competition isn’t Humans vs AI. It’s Humans with AI vs everyone else.

[a16z @a16z]

Narrative violations abound:

- Demand for software engineers is rising
- Software devs are rising as a share of new jobs
- AI exposed industries are seeing above-trend wage growth
- Open PM jobs haven't been higher since 2022

More from a16z's David George on the "AI job apocalypse" myth: https://t.co/7sbadmEElG

[Marty Cornish @martycornish]

Strong agree

[Sawyer Merritt @SawyerMerritt]

Would be good to at least give us the option of selecting routing/navigation as the reason for disengaging, because I’d bet that 80-90% of the time, that is the reason for most people lol.

[Chamath Palihapitiya @chamath]

On page twenty-six of “The Billionaire Tax” proposal in California, it explains how the state legislature can convert from a Billionaire Tax to an Everyone Tax without voter approval.

They can also adjust the tax to be a yearly tax, not just one time…again, without your approval.

Intelligence test for you: if this was meant to just target Billionaires, why did they write this in?

[Hemant Mohapatra @MohapatraHemant]

99.9% of the world is completely unaware what is going to hit them in <5yrs: incredibly powerful, cheap intelligence. The way we are all addicted to information, glued to our screen 24x7, we are about to get addicted to hyper-intelligence. Work is growing 10x from here, unf, not reduce.

[Marty Cornish @martycornish]

@samdblond I’d love to play

[Marty Cornish @martycornish]

@auren

[Marty Cornish @martycornish]

I heard a great line a few years ago from Jeff Weiner (former CEO LinkedIn): “trust equals consistency over time, and there’s no shortcut for either.” While it sounds good, I would argue that healthy conflict does act maybe not to “shortcut” time, but to accelerate it, because it allows people to demonstrate their behavior outside of a normal expectation frame and build stronger trust faster than under unremarkable circumstances. When expectations exist that don’t match reality, breaking those expectations will initially cause resentment. If communication is direct and effective, and both people are flexible enough to change as appropriate to form a new shared reality, that resentment will be short lived, and stronger trust forms on the other side.

[Marty Cornish @martycornish]

@CoachSHoltz Sorry for your loss. The best.

[Naval @naval]

The cost of code is coming down, so we will consume more of it.

The productivity of coders is going up, so they will become more valuable.

Coding now includes training and driving models.

[Marty Cornish @martycornish]

@SawyerMerritt Get 3 months of Full Self-Driving (Supervised) or up to $1,000 off with my referral link. https://t.co/taeL8MKgZo

[Marty Cornish @martycornish]

The other important set of challenges here is 1) of course some individual companies and bureaucrats will make mistakes, 2) if a startup gets caught up in this they will not be able to raise money, 3) most startups do not have the cash or time to allocate to fighting. De facto death sentence for most.

[Garry Tan @garrytan]

The people who wrote the California Asset Seizure Tax say:

“Don’t worry, if the state mis-prices your startup, you can bring in ‘independent appraisers’ to argue with them. And if you can’t pay 5% right now, they’ll put you on a five‑year payment plan. With interest.”

That’s not a safeguard. That’s: we take the money, you fight the DMV on hard mode to get it back.

[david friedberg @friedberg]

California started with the Gold Rush and might end with the Golden Exit.

it has been underreported how much wealth has left CA because of the asset seizure tax being proposed.

a private poll was conducted amongst affected individuals a few days ago and 80-90% surveyed said they have already left CA in 2025 or will leave in 2026 if the ballot measure looks likely to pass.

$2-2.5T of assets gone, representing about $20B of annual revenue for the state government. and likely hundreds of thousands of jobs now at risk.

less reported is the bigger exodus underway from folks who are NOT directly affected but worry (as they should) that this law will quickly transition from billionaires to everyone else...

the initiative actually gives CA legislators the right to take anyone's post-tax assets anytime in the future based on a majority vote. this isn't about billionaires. it's a new "tax system" that simply destroys private property rights in America.

all private property is now public property.
even after paying your taxes, it's not legally your property anymore. it's the government's, you're just borrowing it.
legislators will decide what you get to keep and temporarily use each year.

countless founders, CEOs, and other business leaders are actively looking to move their companies out of state. not just tech, not just AI, not just billionaires, but the core engine of California's prosperity since 1847 is unraveling.

and here is how this initiative risks unraveling America:
- ~10 states have explicit or implicit prohibitions against an asset seizure tax...
- individuals affected in CA (and other states trying to do the same) will move to these states that endow private property rights.
- CA already has a $20-30B annual budget deficit, an unfunded ~$1T pension liability for public employees/unions, and $500B of debt outstanding. the state can not afford to borrow much more and will launch more asset seizures to meet its obligations.
- asset seizures will first transition to "millionaires" and eventually to the entire middle class as more asset seizures drive more people to leave the state.
- the deficit, debt, and job loss will spiral. the Golden Exit.
- no US state has ever declared bankruptcy. in addition to CA, dozens of other states face similar fiscal crises - legislators promised future benefits that can't be paid or theft and waste have been allowed to run rampant and unabated for years.
- struggling states will eventually request federal government assistance, as they always have in times of fiscal crisis, effectively "federalizing state debt".
- states not in crisis will declare "enough is enough", individuals in those states will refuse to pay their federal taxes (why pay for other people's mistakes?), some states may try to secede from the Union, and a constitutional and civil crisis will erupt.

this may seem far-fetched but it is the obvious domino effect of selectively deleting private property rights for some people in some states.

i am not a billionaire and this CA bill does not affect me, but i care about the country and the state of CA. i want both to thrive. it's obvious that there are people in CA in desperate need of support and assistance, and inequities may exist that need to be rectified, but eliminating private property rights is the wrong path for everyone.

a few alternatives to consider first:
1) with a $350B annual budget, CA can cut programs that result in theft and little-to-no benefit for citizens. $50B per year is likely recoverable.
2) if more taxes are needed, tax loans against unrealized capital gains (very few objections will arise), eliminate tax-free rollover of certain appreciated assets (real estate industry will fight), create a step up in basis on inheritance (some will fight but most will support). likely $10Bs of incremental revenue can be realized.
3) restructure all public retirement programs from Defined Benefit to Defined Contribution. eliminating the unfunded retirement liabilities ($1T+) will be the release valve on the future the state so desperately needs.

we must address what ails us without dividing and destroying our state, our nation, our home.

ignore the rhetoric, these are the facts.

[The Boring Company @boringcompany]

Vegas Loop dropoff at the Airport!

Still in the early phases of the program, providing a small number of rides each day.

The ride includes both a tunnel portion and a long surface portion, with the latter shifting to subsurface when the 2.25 mile Airport Connector twin tunnels open, hopefully in Q1 2026.

Thanks to @LASAirport @LVCVA @ClarkCountyNV @NevadaDBI in reaching this milestone.

[Marty Cornish @martycornish]

Cool

[Aakash Gupta @aakashgupta]

The entire robotics industry is about to compress a decade of progress into 18 months, and nobody’s pricing it in.

The hardware has been ready for years. Boston Dynamics had Atlas doing backflips in 2018. The bottleneck was never motors or actuators. It was that every robot behavior had to be hand-coded. Pick up a box? That’s one program. Pick up a bottle? Different program. Move the box from shelf A to shelf B in a warehouse with slightly different lighting? Start over.

Foundation models broke this completely.

Before VLAs, teaching a robot one skill gave you exactly one skill. Zero compounding. Zero transfer. A robot trained to fold shirts couldn’t fold towels without starting from scratch. The labor intensity of data generation meant robotics datasets stayed narrow, robots overfit, and small variations like object weight or table height caused failures.

Now a single Gemini Robotics model handles tasks it has never seen in training. Google’s On-Device model learns new behaviors with 50-100 demonstrations. Not 50,000. Fifty. That’s a 1000x reduction in the data requirement for new capabilities.

The speed implications cascade through everything.

First order: deployment timelines collapse. What took robotics teams 6-12 months of custom programming now takes days of fine-tuning. Second order: the addressable market explodes. Tasks that were never economical to automate suddenly are, because the integration cost dropped by orders of magnitude. Third order: the data flywheel accelerates. Every robot running Gemini Robotics feeds learning back into the foundation model. More deployments means faster improvement means more deployments.

Physical Intelligence raised at $2.4B because investors finally understood this. Boston Dynamics partnered with Toyota Research Institute to bolt Large Behavior Models onto Atlas. Every humanoid company is scrambling to either build or license the intelligence layer they don’t have.

The market is still valuing robotics companies on their hardware differentiation. But hardware is commoditizing. Boston Dynamics spent a decade perfecting locomotion, and now that’s table stakes. The value is migrating entirely to whoever owns the foundation model that generalizes across embodiments.

Google trained Gemini on the largest multimodal corpus ever assembled. Then they added physical actions as an output modality. That’s not a robotics company bolting on AI. That’s an AI company whose models now output motor commands.

The companies pricing this correctly are building around foundation model access, not around proprietary hardware. The companies pricing this wrong are still acting like the moat is in the mechanical engineering.

AGI moving into the physical world isn’t a 10-year prediction. Gemini Robotics shipped in March. The 1.5 version with chain-of-thought reasoning shipped in September. They’re iterating on a 6-month release cycle while hardware companies iterate on 3-year cycles.

The gap between software intelligence timelines and hardware development timelines is the entire trade.

[Marty Cornish @martycornish]

@ericweinstein Awesome Eric. Merry Christmas to you too!

[Marty Cornish @martycornish]

@DavidSacks @Jason @ThomasSowell Bessent did answer this question better. Prices are still high and continue to rise, because reducing inflation, if above zero, doesn’t reduce prices from the already high level they were at. Many are still hurting, even though the overall economic metrics are good.

[David Sacks @DavidSacks]

AI JOB LOSS HOAX EXPOSED

According to a new study from Vanguard, the "occupations most exposed to AI automation are actually outperforming the rest of the labor market in terms of job growth and real wage increases."

Job growth was up 1.7% in occupations with high AI exposure, compared to 0.8% in all other occupations.

Real wages increased 3.8% in occupations with high AI exposure, compared with 0.7% in all other occupations.

Rather than causing job loss, AI is making workers more productive, driving gains in both jobs and wages.

[Marty Cornish @martycornish]

@pmarca @DavidSacks Awesome

[Andrej Karpathy @karpathy]

Sharing an interesting recent conversation on AI's impact on the economy.

AI has been compared to various historical precedents: electricity, industrial revolution, etc., I think the strongest analogy is that of AI as a new computing paradigm (Software 2.0) because both are fundamentally about the automation of digital information processing.

If you were to forecast the impact of computing on the job market in ~1980s, the most predictive feature of a task/job you'd look at is to what extent the algorithm of it is fixed, i.e. are you just mechanically transforming information according to rote, easy to specify rules (e.g. typing, bookkeeping, human calculators, etc.)? Back then, this was the class of programs that the computing capability of that era allowed us to write (by hand, manually).

With AI now, we are able to write new programs that we could never hope to write by hand before. We do it by specifying objectives (e.g. classification accuracy, reward functions), and we search the program space via gradient descent to find neural networks that work well against that objective. This is my Software 2.0 blog post from a while ago. In this new programming paradigm then, the new most predictive feature to look at is verifiability. If a task/job is verifiable, then it is optimizable directly or via reinforcement learning, and a neural net can be trained to work extremely well. It's about to what extent an AI can "practice" something. The environment has to be resettable (you can start a new attempt), efficient (a lot attempts can be made), and rewardable (there is some automated process to reward any specific attempt that was made).

The more a task/job is verifiable, the more amenable it is to automation in the new programming paradigm. If it is not verifiable, it has to fall out from neural net magic of generalization fingers crossed, or via weaker means like imitation. This is what's driving the "jagged" frontier of progress in LLMs. Tasks that are verifiable progress rapidly, including possibly beyond the ability of top experts (e.g. math, code, amount of time spent watching videos, anything that looks like puzzles with correct answers), while many others lag by comparison (creative, strategic, tasks that combine real-world knowledge, state, context and common sense).

Software 1.0 easily automates what you can specify.
Software 2.0 easily automates what you can verify.