Jun 13, 2025

-

Jun 13, 2025

Online

Red Teaming A Narrow Path: ControlAI Policy Sprint

Help stress-test the policies that could prevent human extinction from AI - before they reach lawmakers' desks.

00:00:00:00

00:00:00:00

00:00:00:00

00:00:00:00

Help stress-test the policies that could prevent human extinction from AI - before they reach lawmakers' desks.

This event is ongoing.

This event has concluded.

107

Sign Ups

20

Entries

Overview

Resources

Guidelines

Schedule

Entries

Overview

Arrow

Join us for a critical one-day sprint to red team the policy framework that could determine humanity's future relationship with artificial intelligence. ControlAI has developed "A Narrow Path" - the first comprehensive plan to address extinction risks from Artificial Superintelligence (ASI). These policies are already being actively pushed to lawmakers in the UK and US through their Direct Institutional Plan (DIP), making this red teaming exercise directly relevant to real-world policy implementation.

Your mission: Help strengthen the Phase 0 policies by identifying weaknesses, implementation challenges, and gaps before they reach legislators' desks.

The Challenge We Face

While most AI developments are beneficial, the rise of superintelligent AI (ASI) threatens humanity with extinction. We do not know how to control AI vastly more powerful than us. Should attempts to build superintelligence succeed, this would risk our extinction as a species.

Current AI development is proceeding without adequate safety measures, with reasonable estimates indicating that it could cost only tens to hundreds of billions of dollars to create artificial superintelligence. Meanwhile, the very people developing advanced AI are warning about these risks.

A Narrow Path: The Solution

ControlAI has found no other plan that comprehensively tries to address the issue, so they made one. "A Narrow Path" is structured in three phases:

  • Phase 0: Safety - Immediate policies to prevent ASI development for 20 years

  • Phase 1: Stability - International oversight that doesn't collapse over time

  • Phase 2: Flourishing - Building foundations for safe transformative AI under human control

Real-World Impact

This isn't theoretical policy research. ControlAI launched a pilot campaign focused on UK lawmakers that validated their approach. In less than three months, over 20 cross-party UK parliamentarians publicly supported their campaign. They succeeded in gaining support in 1 out of every 3 cases when briefing lawmakers.

Recent polling shows that a large majority (74%) of Brits support placing the UK's AI Safety Institute on a statutory footing, and 16 British lawmakers have signed a statement calling for new AI laws targeted specifically at "superintelligent" AI systems.

Focus of This Sprint

We're exclusively red teaming Phase 0 policies because:

  1. These are the policies actively being pushed through the DIP

  2. They are the necessary first step - stopping ASI development is the precondition for everything else

  3. Real-world implementation is imminent, making your feedback immediately actionable

Prizes

1st Place: $300 + exclusive mentoring session with Control AI's policy team to discuss your findings and potential integration into their Direct Institutional Plan advocacy efforts.

2nd Place: $200 + mentoring session with Control AI's policy team to explore how your red teaming insights could strengthen real-world policy implementation.

107

Sign Ups

20

Entries

Overview

Resources

Guidelines

Schedule

Entries

Overview

Arrow

Join us for a critical one-day sprint to red team the policy framework that could determine humanity's future relationship with artificial intelligence. ControlAI has developed "A Narrow Path" - the first comprehensive plan to address extinction risks from Artificial Superintelligence (ASI). These policies are already being actively pushed to lawmakers in the UK and US through their Direct Institutional Plan (DIP), making this red teaming exercise directly relevant to real-world policy implementation.

Your mission: Help strengthen the Phase 0 policies by identifying weaknesses, implementation challenges, and gaps before they reach legislators' desks.

The Challenge We Face

While most AI developments are beneficial, the rise of superintelligent AI (ASI) threatens humanity with extinction. We do not know how to control AI vastly more powerful than us. Should attempts to build superintelligence succeed, this would risk our extinction as a species.

Current AI development is proceeding without adequate safety measures, with reasonable estimates indicating that it could cost only tens to hundreds of billions of dollars to create artificial superintelligence. Meanwhile, the very people developing advanced AI are warning about these risks.

A Narrow Path: The Solution

ControlAI has found no other plan that comprehensively tries to address the issue, so they made one. "A Narrow Path" is structured in three phases:

  • Phase 0: Safety - Immediate policies to prevent ASI development for 20 years

  • Phase 1: Stability - International oversight that doesn't collapse over time

  • Phase 2: Flourishing - Building foundations for safe transformative AI under human control

Real-World Impact

This isn't theoretical policy research. ControlAI launched a pilot campaign focused on UK lawmakers that validated their approach. In less than three months, over 20 cross-party UK parliamentarians publicly supported their campaign. They succeeded in gaining support in 1 out of every 3 cases when briefing lawmakers.

Recent polling shows that a large majority (74%) of Brits support placing the UK's AI Safety Institute on a statutory footing, and 16 British lawmakers have signed a statement calling for new AI laws targeted specifically at "superintelligent" AI systems.

Focus of This Sprint

We're exclusively red teaming Phase 0 policies because:

  1. These are the policies actively being pushed through the DIP

  2. They are the necessary first step - stopping ASI development is the precondition for everything else

  3. Real-world implementation is imminent, making your feedback immediately actionable

Prizes

1st Place: $300 + exclusive mentoring session with Control AI's policy team to discuss your findings and potential integration into their Direct Institutional Plan advocacy efforts.

2nd Place: $200 + mentoring session with Control AI's policy team to explore how your red teaming insights could strengthen real-world policy implementation.

107

Sign Ups

20

Entries

Overview

Resources

Guidelines

Schedule

Entries

Overview

Arrow

Join us for a critical one-day sprint to red team the policy framework that could determine humanity's future relationship with artificial intelligence. ControlAI has developed "A Narrow Path" - the first comprehensive plan to address extinction risks from Artificial Superintelligence (ASI). These policies are already being actively pushed to lawmakers in the UK and US through their Direct Institutional Plan (DIP), making this red teaming exercise directly relevant to real-world policy implementation.

Your mission: Help strengthen the Phase 0 policies by identifying weaknesses, implementation challenges, and gaps before they reach legislators' desks.

The Challenge We Face

While most AI developments are beneficial, the rise of superintelligent AI (ASI) threatens humanity with extinction. We do not know how to control AI vastly more powerful than us. Should attempts to build superintelligence succeed, this would risk our extinction as a species.

Current AI development is proceeding without adequate safety measures, with reasonable estimates indicating that it could cost only tens to hundreds of billions of dollars to create artificial superintelligence. Meanwhile, the very people developing advanced AI are warning about these risks.

A Narrow Path: The Solution

ControlAI has found no other plan that comprehensively tries to address the issue, so they made one. "A Narrow Path" is structured in three phases:

  • Phase 0: Safety - Immediate policies to prevent ASI development for 20 years

  • Phase 1: Stability - International oversight that doesn't collapse over time

  • Phase 2: Flourishing - Building foundations for safe transformative AI under human control

Real-World Impact

This isn't theoretical policy research. ControlAI launched a pilot campaign focused on UK lawmakers that validated their approach. In less than three months, over 20 cross-party UK parliamentarians publicly supported their campaign. They succeeded in gaining support in 1 out of every 3 cases when briefing lawmakers.

Recent polling shows that a large majority (74%) of Brits support placing the UK's AI Safety Institute on a statutory footing, and 16 British lawmakers have signed a statement calling for new AI laws targeted specifically at "superintelligent" AI systems.

Focus of This Sprint

We're exclusively red teaming Phase 0 policies because:

  1. These are the policies actively being pushed through the DIP

  2. They are the necessary first step - stopping ASI development is the precondition for everything else

  3. Real-world implementation is imminent, making your feedback immediately actionable

Prizes

1st Place: $300 + exclusive mentoring session with Control AI's policy team to discuss your findings and potential integration into their Direct Institutional Plan advocacy efforts.

2nd Place: $200 + mentoring session with Control AI's policy team to explore how your red teaming insights could strengthen real-world policy implementation.

107

Sign Ups

20

Entries

Overview

Resources

Guidelines

Schedule

Entries

Overview

Arrow

Join us for a critical one-day sprint to red team the policy framework that could determine humanity's future relationship with artificial intelligence. ControlAI has developed "A Narrow Path" - the first comprehensive plan to address extinction risks from Artificial Superintelligence (ASI). These policies are already being actively pushed to lawmakers in the UK and US through their Direct Institutional Plan (DIP), making this red teaming exercise directly relevant to real-world policy implementation.

Your mission: Help strengthen the Phase 0 policies by identifying weaknesses, implementation challenges, and gaps before they reach legislators' desks.

The Challenge We Face

While most AI developments are beneficial, the rise of superintelligent AI (ASI) threatens humanity with extinction. We do not know how to control AI vastly more powerful than us. Should attempts to build superintelligence succeed, this would risk our extinction as a species.

Current AI development is proceeding without adequate safety measures, with reasonable estimates indicating that it could cost only tens to hundreds of billions of dollars to create artificial superintelligence. Meanwhile, the very people developing advanced AI are warning about these risks.

A Narrow Path: The Solution

ControlAI has found no other plan that comprehensively tries to address the issue, so they made one. "A Narrow Path" is structured in three phases:

  • Phase 0: Safety - Immediate policies to prevent ASI development for 20 years

  • Phase 1: Stability - International oversight that doesn't collapse over time

  • Phase 2: Flourishing - Building foundations for safe transformative AI under human control

Real-World Impact

This isn't theoretical policy research. ControlAI launched a pilot campaign focused on UK lawmakers that validated their approach. In less than three months, over 20 cross-party UK parliamentarians publicly supported their campaign. They succeeded in gaining support in 1 out of every 3 cases when briefing lawmakers.

Recent polling shows that a large majority (74%) of Brits support placing the UK's AI Safety Institute on a statutory footing, and 16 British lawmakers have signed a statement calling for new AI laws targeted specifically at "superintelligent" AI systems.

Focus of This Sprint

We're exclusively red teaming Phase 0 policies because:

  1. These are the policies actively being pushed through the DIP

  2. They are the necessary first step - stopping ASI development is the precondition for everything else

  3. Real-world implementation is imminent, making your feedback immediately actionable

Prizes

1st Place: $300 + exclusive mentoring session with Control AI's policy team to discuss your findings and potential integration into their Direct Institutional Plan advocacy efforts.

2nd Place: $200 + mentoring session with Control AI's policy team to explore how your red teaming insights could strengthen real-world policy implementation.

Speakers & Collaborators

Adam Shimi

Organiser

Epistemology researcher at Control AI developing policy frameworks to address superintelligence risks, with expertise in technical writing and AI safety theory.


Max Winga

Organiser

Communications analyst at Control AI focused on preventing artificial superintelligence extinction threats through strategic messaging and public engagement

Tolga Bilge

Judge

Policy researcher and superforecaster at ControlAI, co-author of "A Narrow Path," with expertise in AI risk assessment and coalition building for international AI governance.

Dave Kasten

Judge

Independent consultant specializing in AI policy, national security strategy, and organizational risk management, with extensive experience in cybersecurity and data protection.

Andrea Miotti

Keynote Speaker

Founder and Executive Director, ControlAI and Advisor at Conjecture

Speakers & Collaborators

Adam Shimi

Organiser

Epistemology researcher at Control AI developing policy frameworks to address superintelligence risks, with expertise in technical writing and AI safety theory.


Max Winga

Organiser

Communications analyst at Control AI focused on preventing artificial superintelligence extinction threats through strategic messaging and public engagement

Tolga Bilge

Judge

Policy researcher and superforecaster at ControlAI, co-author of "A Narrow Path," with expertise in AI risk assessment and coalition building for international AI governance.

Dave Kasten

Judge

Independent consultant specializing in AI policy, national security strategy, and organizational risk management, with extensive experience in cybersecurity and data protection.

Andrea Miotti

Keynote Speaker

Founder and Executive Director, ControlAI and Advisor at Conjecture

Registered Jam Sites

Register A Location

Beside the remote and virtual participation, our amazing organizers also host local hackathon locations where you can meet up in-person and connect with others in your area.

The in-person events for the Apart Sprints are run by passionate individuals just like you! We organize the schedule, speakers, and starter templates, and you can focus on engaging your local research, student, and engineering community.

We haven't announced jam sites yet

Check back later

Registered Jam Sites

Register A Location

Beside the remote and virtual participation, our amazing organizers also host local hackathon locations where you can meet up in-person and connect with others in your area.

The in-person events for the Apart Sprints are run by passionate individuals just like you! We organize the schedule, speakers, and starter templates, and you can focus on engaging your local research, student, and engineering community.

We haven't announced jam sites yet

Check back later