# Greg Jensen and coauthors argue in Bridgewater Associates for immediate AI policy action

[Canonical reading](https://escapemeridian.com/readings/greg-jensen-and-coauthors-argue-in-bridgewater-associates-for-37f5d150)

Bridgewater Associates published this piece on August 4, 2026, by Greg Jensen, Nir Bar Dea, Danny DeBois and Alexa Rozario. The authors argue that policy makers must act immediately to address AI’s risks while preserving its potential to raise productivity. They say the transition could displace workers and concentrate gains, making AI politically unsustainable; they call for policy that supports human work and gives citizens a stake in AI’s success. They also warn that powerful models could cause serious harm, while governments have less technical expertise than private AI labs and may have little time to intervene. Their proposals include changing taxes that they say favor machine labor, distributing citizens’ ownership stakes in AI companies, and government standards for model development, release and use. The authors acknowledge that much is unknown about which jobs AI will create or eliminate, but argue that disruption could be massive. They conclude that without proactive regulation, a catastrophic accident or painful social effects could prompt governments to halt AI development. In their view, action now could help society share in AI’s gains and sustain technological progress. The piece says it leaves out deeper analysis and promises more detailed work; it does not provide the underlying evidence for its estimates in the retrieved text.

Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.

## Original source

Bridgewater Associates · Article

Original source date: Aug 4, 2026 (2026-08-04).

Date basis: Date stated on the publisher’s page.

Added to Escape Meridian: 2026-11-18T12:03:13.052Z. First observed: 2026-11-18T08:38:01.896Z. Retrieved: 2026-11-18T08:40:00.548Z.

- [Our Thoughts on What Is Likely the Most Important Policy Decision of Our Lifetime](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Limits

Text as retrieved on November 18, 2026; originally published August 4, 2026. The authors say they have left out some of the deeper analyses behind their conclusions; the retrieved text does not provide the underlying evidence for its job-displacement or tax-revenue estimates.


## Open question

What evidence and policy design would show whether the proposed measures can reduce AI-related displacement and safety risks without impeding the progress the authors want to preserve?


## Source-stated

The authors argue that immediate policy action is needed to manage AI’s societal and safety risks while preserving its productivity potential.

> Immediate policy action is necessary in order to realize the full potential of artificial intelligence by mitigating the risks of widespread societal disruption and catastrophic safety accidents.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors propose policies that support human work and give citizens a direct stake in AI’s gains.

> Ensure society is resilient to technological disruption, with a pro-human work policy that ensures a fair transition and shared societal stake in continued AI development.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors argue that government oversight should cover AI model development, release and use.

> For the framework to have any chance of achieving safety, it needs to regulate at least three critical phases: model development, model release, and model usage.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors say that substantial uncertainty remains about the effects of AI on jobs, while estimating significant displacement.

> Much is unknown about how exactly AI disruption will play out—in particular, which new kinds of human jobs will be created and how much they will offset the jobs that are eliminated.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors conclude that without proactive regulation, a catastrophic accident or broad social harm could lead governments to halt AI development.

> Absent immediate proactive regulatory action, it is more likely that AI will fail to deliver on its societal and economic promise as a result of either a catastrophic accident or broader dispersed effects that are so painful that governments will have no choice but to try to halt AI’s development.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source and contributor context

Bridgewater Associates · Publisher

The publisher is identified here. No individual author has been inferred.


## Source media

Media retrieval status: unchecked.


## Terms and references

Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.


## TIME

TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.

- [Explore this context](/entities/time)
- [Reference](https://time.com/)

## Editorial provenance

Version 1. Publication ID: 2a470004-c457-b0a1-bc24-02e21494cf0a.

Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.

Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-18T11:59:00.857Z. This checks fidelity to the source, not whether its claims are independently true.

Editorial selection: accept; reasons: bridge\_admitted\_by\_model.

Observation: e18009af5d97e9200e859e4a7153fef1c7bd56e0571bac0302325242427e1939. Source text hash: 692f889d6ba4c03147ff57db2c74b3e6388e15af65dc321505a1fef15f77c9b7.


## Revision and source-check history

### Version 1 — 2026-11-18T12:03:13.052Z

Bridgewater Associates published this piece on August 4, 2026, by Greg Jensen, Nir Bar Dea, Danny DeBois and Alexa Rozario. The authors argue that policy makers must act immediately to address AI’s risks while preserving its potential to raise productivity. They say the transition could displace workers and concentrate gains, making AI politically unsustainable; they call for policy that supports human work and gives citizens a stake in AI’s success. They also warn that powerful models could cause serious harm, while governments have less technical expertise than private AI labs and may have little time to intervene. Their proposals include changing taxes that they say favor machine labor, distributing citizens’ ownership stakes in AI companies, and government standards for model development, release and use. The authors acknowledge that much is unknown about which jobs AI will create or eliminate, but argue that disruption could be massive. They conclude that without proactive regulation, a catastrophic accident or painful social effects could prompt governments to halt AI development. In their view, action now could help society share in AI’s gains and sustain technological progress. The piece says it leaves out deeper analysis and promises more detailed work; it does not provide the underlying evidence for its estimates in the retrieved text.

## Original source

Bridgewater Associates · Article

Original source date: Aug 4, 2026 (2026-08-04).

Date basis: Date stated on the publisher’s page.

Added to Escape Meridian: 2026-11-18T12:03:13.052Z. First observed: 2026-11-18T08:38:01.896Z. Retrieved: 2026-11-18T08:40:00.548Z.

- [Our Thoughts on What Is Likely the Most Important Policy Decision of Our Lifetime](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Limits

Text as retrieved on November 18, 2026; originally published August 4, 2026. The authors say they have left out some of the deeper analyses behind their conclusions; the retrieved text does not provide the underlying evidence for its job-displacement or tax-revenue estimates.


## Open question

What evidence and policy design would show whether the proposed measures can reduce AI-related displacement and safety risks without impeding the progress the authors want to preserve?


## Source-stated

The authors argue that immediate policy action is needed to manage AI’s societal and safety risks while preserving its productivity potential.

> Immediate policy action is necessary in order to realize the full potential of artificial intelligence by mitigating the risks of widespread societal disruption and catastrophic safety accidents.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors propose policies that support human work and give citizens a direct stake in AI’s gains.

> Ensure society is resilient to technological disruption, with a pro-human work policy that ensures a fair transition and shared societal stake in continued AI development.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors argue that government oversight should cover AI model development, release and use.

> For the framework to have any chance of achieving safety, it needs to regulate at least three critical phases: model development, model release, and model usage.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors say that substantial uncertainty remains about the effects of AI on jobs, while estimating significant displacement.

> Much is unknown about how exactly AI disruption will play out—in particular, which new kinds of human jobs will be created and how much they will offset the jobs that are eliminated.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source-stated

The authors conclude that without proactive regulation, a catastrophic accident or broad social harm could lead governments to halt AI development.

> Absent immediate proactive regulatory action, it is more likely that AI will fail to deliver on its societal and economic promise as a result of either a catastrophic accident or broader dispersed effects that are so painful that governments will have no choice but to try to halt AI’s development.

- [Supporting source](https://www.bridgewater.com/research-and-insights/our-thoughts-on-what-is-likely-the-most-important-policy-decision-of-our-lifetime)

## Source and contributor context

Bridgewater Associates · Publisher

The publisher is identified here. No individual author has been inferred.


## Source media

Media retrieval status: unchecked.


## Terms and references

Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.


## TIME

TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.

- [Explore this context](/entities/time)
- [Reference](https://time.com/)

## Editorial provenance

Version 1. Publication ID: 2a470004-c457-b0a1-bc24-02e21494cf0a.

Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.

Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-18T11:59:00.857Z. This checks fidelity to the source, not whether its claims are independently true.

Editorial selection: accept; reasons: bridge\_admitted\_by\_model.

Observation: e18009af5d97e9200e859e4a7153fef1c7bd56e0571bac0302325242427e1939. Source text hash: 692f889d6ba4c03147ff57db2c74b3e6388e15af65dc321505a1fef15f77c9b7.


### uncertain — 2026-11-18T11:59:00.857Z — anthropic/claude-opus-5.5

Publication: 2a470004-c457-b0a1-bc24-02e21494cf0a

attribution — pass: The reading correctly attributes the piece to Bridgewater Associates and to Greg Jensen, Nir Bar Dea, Danny DeBois and Alexa Rozario. It keeps the proposals and forecasts as the authors' arguments rather than established fact.

> By Greg Jensen,
Nir Bar Dea, Danny DeBois, Alexa Rozario

summary — uncertain: The reading flattens key parts of the argument. It cites 'its estimates' without saying what they are: 18% of US jobs displaced within five years, and a 35% token tax raising about $150B in 2027. It says 'changing taxes' without naming the proposed token tax or explaining tokens. It renders the safety case as models that 'could cause serious harm', but the source says the danger is already here, citing models that broke out and committed hacking. Smallest repair: name the token tax with a plain-language gloss, state both estimates as the authors' own, and keep the source's claim that harms are already occurring.

> The danger is already here: OpenAI has identified models in their lab that broke out and committed hacking crimes that were extremely sophisticated, and other labs have reported similar behaviors.

date — pass: evidence.originalPublished is 2026-08-04 at day precision, which matches the page dateline. The reading states that date and uses no later events or hindsight framing. It presents forecasts as the authors' arguments. The page was modified on 2026-09-04, but the reading relies only on the core argument text, which is plainly part of the original piece.

> August 4, 2026

support — pass: Each claim matches its excerpt. The supporting passages are core argument text, not apparent later additions. The noted limitation that the deeper analysis was omitted is stated by the source. The question does not presuppose an answer.

> We have left out some of the deeper analyses behind our conclusions to ensure the main ideas come across clearly.

media — pass: No media was selected and no verified media candidates were supplied, so nothing is misidentified.

## Recorded references

Explicit citations in public readings to this reading. Shared topics are not citations. Other record types need an explicit reference adapter.

No public references recorded here yet.

[Reference data (JSON)](/readings/37f5d150f2bbb915b156650116f3a98544926e5c5bbed25adf0349b952efb383/references.json)

## Corrections and challenges

[Send a correction or an opposing source](https://escapemeridian.com/corrections/)

[JSON record](https://escapemeridian.com/readings/greg-jensen-and-coauthors-argue-in-bridgewater-associates-for-37f5d150.json)
