ACL Caregiver AI Prize Challenge (Phase 1, 2026–2029)

ACL Caregiver AI Prize Challenge (Phase 1, 2026–2029)

The ACL Caregiver AI Prize Challenge is a federal prize competition run under the Administration for Community Living (ACL), part of HHS. It is intended to support AI tools that reduce burden on family caregivers and the paid caregiving workforce, improve quality of care, and strengthen the longer-term caregiving infrastructure. Unlike a classic grant solicitation that asks for a single institution-driven proposal, this is a prize format with submission gates, phase-based progression, and practical deliverable expectations.

The official listing on USAGov shows an open window from 2026-02-06 9:17 AM ET to 2029-02-06 5:00 PM ET and lists total cash prizes of $2,500,000. The ACL/HHS communication confirms a three-phase structure and a Phase 1 budget envelope of up to $2.5M, with up to 20 winners expected in that first wave.

For teams that want to start now, this is still useful because the competition has a broad multi-year timeline and explicit phase milestones. Even if you missed an earlier pre-application step, you still need to monitor the challenge dashboard and official challenge page for updated status.

Key details

What this opportunity actually offers

This is not a general-purpose technology grant for a broad portfolio. The competition is narrowly focused on caregiving outcomes, particularly practical improvements at points where care teams and families are overloaded. Challenge language emphasizes:

• reducing caregiver burden,
• supporting quality at home and in community care,
• and enabling safer, more efficient workflows.

This means your project should not be framed only as a clever AI experiment. It should be structured as a practical intervention in real care workflows.

It is important to distinguish what this program does and does not do:

• It is not a routine R&D award that automatically funds baseline pilots.
• It is not restricted to one grant category like SBIR or a federal contract.
• It is structured as a competition with winners receiving prizes after evaluation.

A strong team should interpret this as a “demonstrate and prove usefulness” mechanism more than “funding as planning support.” In other words, your value proposition must show what changes for caregivers and how it can be deployed ethically and reliably.

How the challenge is organized

The program is organized as a staged process with three tracks and phase progression logic:

1. Application route through Phase 1.
2. Eligibility and award rules.
3. Phased prize eligibility in later rounds.

Publicly available summaries mention a three-phase model. Phase 1 was described as the first practical entry lane, with a broader prize envelope and category-based recognition. The program also indicates that teams move forward across phases through performance, not by promising capability alone.

Why this matters for applicants

In a conventional grant process, reviewers often focus on whether your written plan is scientifically plausible and administratively complete. Here, the competition dynamic adds a second axis: whether your solution is strong enough to be judged by implementation and relevance criteria tied to caregiver outcomes and safety.

This has three consequences:

• Execution matters as much as idea quality. A vague but brilliant concept is unlikely to stand out.
• Evidence planning is mandatory. You must describe how AI helps caregivers in real use contexts.
• Your proposal should map to track selection. A solution can accidentally dilute impact if it blends both tracks without clear operational scope.

Who this is for and who should skip it

This challenge is useful for teams that can align technical depth with care realism.

Best fit

• AI developers with healthcare-adjacent products.
• startups or product teams with user research in caregiving contexts.
• care service design teams that can co-design with families and workforce organizations.
• researchers and social entrepreneurs already working where aging, disability support, and home care interact.

Better not to apply if you are

• mainly at an idea-only stage and do not have a realistic implementation path,
• unable to commit to clear data handling, privacy, or safety assumptions,
• planning an abstract machine-learning demo with no real user setting,
• or expecting long lead-time grant funding and compliance simplicity.

Because of the phased structure, teams that lack an MVP, user testing, or partner context should use this phase to prototype quickly and avoid overbuilding narrative around future partnerships.

Eligibility and participation rules that influence your filing

Publicly visible materials for the challenge include explicit practical requirements for participation and submission formatting. While some operational lines are on ACL’s own challenge page, the official summary and associated HHS communication make several constraints clear.

At a minimum, your filing should account for:

• Who submits: individual, entity, or team participation.
• Primary point of contact: the named lead is expected to be responsible for all communication.
• Eligibility posture: underage applicants need guardian authorization.
• Submission format: English-language materials, PDF or Word format, with page-length and margin constraints for the formal application template.

The 2026 challenge communications also highlight important practical constraints:

• Use a format aligned with the official application outline and phase expectations.
• Keep materials concise and reviewable.
• For US recipients, the lead role for payments and award follow-through needs to be clear.
• Federal grantees should verify whether spending challenge development time with grant funds is permitted under their specific award terms.

Practical interpretation of eligibility

For this kind of program, eligibility is less about formal degree checkboxes and more about alignment of roles:

• Make sure at least one accountable lead is clearly identified.
• Ensure your team can answer questions on safety, fairness, and implementation.
• Keep partner roles explicit (especially if you blend software and service workflows).

If your team has a non-U.S. entity without a clear U.S. lead in Phase 1, do not assume this is impossible, but do assume this needs stronger planning around compliance and payout rules.

Application timeline and sequence (as known)

The challenge page family is part of the federal competition ecosystem where the published listing and agency communications show different levels of detail.

A useful reading of the current state at 2026-05-31:

• Start window: listed as 2026-02-06.
• Phase 1 intent / early route: publicly encouraged by agency briefings to submit a short intent early in the phase.
• Phase 1 full application: stated due around 2026-07-31 5:00 PM ET.
• Grand window on the USAGov summary: through 2029, indicating a multi-year structure.

Because this is a federal challenge, deadlines can shift in later updates. Treat any specific dates as valid only after re-checking the linked official challenge page around submission time.

Suggested workflow from discovery to submission

1. Week 1: Track decision + use-case fit

   • Choose whether you are applying for family support tools or workforce support tools.
   • Write a one-page impact statement linked to either caregiver burden or workflow reliability.

2. Week 2: Problem evidence bundle

   • Collect user interviews or letters of relevance.
   • Define measurable outcomes: time saved, burden reduced, error reduction, scheduling uplift, communication latency, etc.

3. Week 3: Architecture + compliance alignment

   • Define data flows, user safeguards, fallback behavior, and privacy boundaries.
   • Verify AI outputs are explainable enough for caregivers and supervisors.

4. Week 4: Submission-ready draft

   • Use required page limits and formatting rules.
   • Keep language plain and operational.

5. Submission period: pre-flight

   • Re-check files for formatting and naming.
   • Confirm one-point contact and signature responsibilities.

How reviewers likely read your application

For this challenge, reviewers are looking for utility plus evidence. A strong proposal should show:

• A narrow, urgent use-case: family caregiver burden, documentation load, training gaps, home care communication failure, or workforce scheduling friction.
• Feasibility of deployment: pilots, MVP quality, and integration readiness for care environments.
• Responsible AI posture: privacy, controllability, human-in-the-loop accountability.
• User-centered design: input from caregivers and home care staff rather than only technical benchmarking.
• Clear scaling trajectory: if it works with one care team, how will it scale?

In many government prize contexts, vague claims are punished more than imperfect but honest proposals. If you do not yet have full production-grade performance, do not hide that; show a realistic validation roadmap.

Application materials checklist (what to prepare now)

A practical submission bundle should include at least:

• concise project summary and problem framing,
• target user profiles (family caregiver / direct care workforce),
• implementation architecture summary,
• safety and privacy controls,
• measurable success criteria,
• timeline with milestones,
• and funding-use explanation tied to build, testing, and deployment readiness.

From the challenge communication, your final package should be easy for reviewers to score quickly. Keep sections short, consistent, and directly tied to challenge priorities:

• Caregiver support outcomes (what changes for the user),
• Technical feasibility (how it runs in real workflows),
• Responsible design (how misuse and bias are reduced),
• Implementation realism (who can pilot and by when).

If an application expects full deployment at submission, this is usually a mismatch. Show a credible path to early field utility instead.

Common mistakes and failure patterns

The most frequent errors in challenge applications are repeatable:

1. Applying to both tracks without a clear lead concept.

   • Fix: pick one primary user path and keep the solution statement narrow.

2. Overpromising AI capabilities.

   • Fix: avoid “universal solution” language; state specific contexts and boundaries.

3. Ignoring responsible AI criteria in design.

   • Fix: include human oversight and user trust safeguards, even if brief.

4. Wrong file structure or missing submission details.

   • Fix: follow page length and format limits exactly.

5. Weak lead ownership.

   • Fix: appoint one point of contact and provide clear team responsibilities.

6. No care setting evidence.

   • Fix: include direct examples from caregiver interviews, observations, or pilot outputs.

Each of these failures is usually preventable with a short pre-submission checklist.

FAQ for this opportunity

Is this open to non-U.S. teams?

The competition is publicly accessible in principle to teams, but payout and eligibility pathways may require U.S. lead responsibilities and compliant documentation.

Is this a research grant?

It is framed as a federal prize challenge, not a traditional grant line with broad discretion. Prize payout and participation rules matter more than legacy grant assumptions.

Is the deadline still open now?

As of the listed records, Phase 1 full submission is around late July 2026. Always confirm current official pages before final submission, because federal challenge pages can post schedule updates.

Are only AI startups eligible?

No. The competition invites a broad set of teams that can deliver practical AI-enabled caregiving outcomes. A cross-functional coalition with strong caregiving operations knowledge can be stronger than a pure software shop.

Do winners get paid directly by ACL?

Public materials indicate direct award delivery and tax handling through standard federal award mechanisms. Payment structure and responsibility follow ACL/HHS instructions.

Strategic fit for 2026/2027 cycle planning

If you are targeting 2026/2027 opportunities, this challenge is one of the few federal AI-caregiving programs with a clear practical social-impact goal and a staged competition model. It can support teams at different maturity levels:

• early teams: proof-of-concept and user-centered framing,
• execution teams: robust piloting, implementation plan, and safety design,
• scale teams: integration strategy with existing care ecosystems.

A practical plan for this cycle is not to treat this as a single “send one email” exercise. Treat it as a staged build-and-prove workflow with periodic public-status checks.

Official links and source pages

• USAGov challenge entry summary: https://www.usa.gov/challenges/acl-caregiver-ai-prize
• ACL challenge landing page: https://acl.gov/caregiver-ai-challenge
• HHS launch note with phase framing and challenge priorities: https://www.hhs.gov/press-room/acl-launches-phase-1-caregiver-ai-prize-competition.html

The most reliable strategy is to use the official challenge page for final submission instructions, then submit from the required contact method with the exact formatting rules and time window stated there.