Project summary

Copied from https://app.grantmaking.ai/projects/ff2c100d-9bd4-46cc-b7e7-e66e844b1fca. Scope changed after original submission: one volunteer will attend rather than two, no director will attend, and expected spend is roughly US$3,000 of the US$5,000.

What this project is

PauseAI Australia got founded last year and has been run by volunteers ever since. We coordinate with PauseAI Global and other national chapters.

The core team of 7 regular volunteers is putting in ~40 hours a week total, yet has no advocacy background and is self-taught. We are self-funded and have received no grants.

Sending two volunteers to PauseCon to learn from the best will let them teach remaining volunteers the skills, and expand PauseAI Australia to affect Australia's policy. At least one member will be an existing director. Ideal funding lets us send three volunteers.

Volunteers will within 4 weeks of PauseCon deliver a training session available to all Australian volunteers, and book at least 10 MP meetings within the following 6 months.

Theory of impact

Global treaties are a coordination problem. While the most impactful countries are the US and China, middle powers can form coalitions and nudge the superpowers onto a safer trajectory for AI development. This was the topic of many sessions at the recent Australian AI Safety Forum 2026.

Sent volunteers will learn the playbook from PauseAI UK and other attending organisations that are more mature, then we teach other volunteers in Australia.

We have met with >20 Federal and state politicians across the major parties, on zero budget. We referred OpenAI to the Australian Federal Police and Attorney General, which got coverage in the ACS's Information Age, a national tech industry outlet. Our media outreach led to 4 other news articles. We coordinated 5 Senator signatures to the Superintelligence Statement. We advocated for over a year for Australia to have an AISI, and for it to focus on existential risk. The AISI has now been launched, and one of its three stated aims is catastrophic and existential risk.

A social movement is essential for x-risk policy change (lesswrong.com/posts/aoqhszdEWqcFWbnda/existential-ai-safety-needs-an-effective-social-movement) and we are the core of that movement in Australia.

The alternative to receiving this grant is that we do not attend PauseCon.

How the money will be spent

Minimum funding $5000 USD

Return economy air fares from Australia to London at current prices = ~$1.7K x 2 seats = ~3.4K USD

Shared accommodation for 5 nights = ~$100 x 5 nights = ~$500 USD

(PauseCon offers free 3-night accommodation but has limited spots)

Metro transport + incidentals + meals = ~$100 USD

Changes in airfares between now and grant allocation <= $1000 USD

Ideal funding $7500 USD

Extra volunteer air fare = ~$1.7K current fare + $500 potential increase = ~2.2K final airfare

Extra 2 nights in London to meet with PauseAI UK and ControlAI people = ~$100 x 2 nights = $200 USD

Potentially larger room(s) to fit 3 people = $100 USD

We are happy to return any unused funds (e.g. due to air fare prices increasing minimally, or us receiving PauseCon accommodation on some of the days).

Who is on your team? What's your track record on similar projects? (includes updates since July 2026)

Entirely volunteer run: 10 in the core team, being four directors and six other regular volunteers doing ~45 hours a week between us, with organisers in Canberra, Melbourne, Sydney, Brisbane and Perth. (Added three volunteers since grantmaking.ai submission in early July).

I lead the grant: director since July 2025, with a background in enterprise automation. The PauseCon attendee organises our Perth group, and has a background in enterprise risk.

Monthly progress: pauseai.info/australia-progress Track record summarised in Theory Of Impact above.

What are the most likely causes and outcomes if this project fails?

Single attendee, illness / travel disruption means nobody from Australian team attends PauseCon.

Alternatively there is issue transmitting training from attendee to wider chapter. Mitigated by recording training for all volunteers, sharing learnings in monthly calls. The total of 10 MP meetings could vary in quality and are an indicative metric rather than the end goal.

How much money have you raised in the last 12 months, and from where?

No grants. Self-funded since founding. One volunteer has donated AU$100 a week since January 2026, with no restrictions or reporting requirements. I put in a thousand dollars of my own. Volunteers otherwise cover their own costs. This US$5000 is our first grant.