Overview

AI Trailblazers is a joint initiative by the Singapore government (MCI, DISG, SNDGO) and Google Cloud, aimed at accelerating generative AI adoption in Singapore across public agencies and private sector organizations. The core purpose is to help organisations identify generative AI use cases, prototype/experiment using Google Cloud's AI toolsets, and mature them toward production in a sandboxed, low-risk environment. It includes providing Innovation Sandboxes, technical support/workshops, access to Google Cloud AI infrastructure/tools, and evaluation/recognition of prototype solutions.

Support Details

Support Rate

No cost/zero fee during sandbox phase - access to infrastructure and tools provided at no cost for prescribed duration

Cap

AI Trailblazers 2.0: Up to 150 more organizations, 10-week sandbox duration. Successor AI CTO programme: Up to S$500,000 incentives

Project Duration & Scope

Initial phase: 3 months sandbox access, AI Trailblazers 2.0: Up to 10 weeks, Successor AI CTO: Longer-term scaling support

Eligibility Criteria

Both public agencies and Singapore-based companies/organisations are eligible
Must be able to propose real-world generative AI use cases (problems to solve via gen AI)
Willing to prototype in the sandbox environment
Should have some internal capacity (developers, AI/data engineers) to engage in prototyping
Workshops aimed at AI practitioners within those organisations
Adherence to programme rules (sandbox duration, evaluation, intellectual property, responsible AI usage)
Expression of interest/application submission to be accepted into a cohort

Supported Categories

Innovation Sandboxes: Free access to Google Cloud's AI infrastructure (GPUs, Vertex AI, foundation models, low-code developer tools) for building prototypes

Workshops & Training: Mandatory full-day workshops/bootcamps for AI practitioners in ideation, use case selection, and prototyping with generative AI

Technical Support: Google Cloud engineers/partners provide technical check-ins, mentoring, guidance to overcome prototype challenges

Recognition & Capstone Projects: High potential MVPs elevated to Capstone Projects, showcased or further supported/incubated

Application Process

Express interest via official channels (EDB/DISG/programme website) as cohorts open

Submit expression of interest/application with proposed generative AI use cases

Be selected into an innovation sandbox cohort through evaluation process

Participate in mandatory workshops, prototyping, and follow programme schedule

Build prototype in sandbox environment using Google Cloud AI tools

Present MVP to panel of evaluators (government + Google) for assessment

Possible elevation to Capstone Projects for high potential MVPs

Transition consideration for successor programmes like AI CTO/AI Cloud Takeoff

Processing Time

Cohort-based programme with quarterly intakes, Sandbox duration: 3 months (initial) to 10 weeks (2.0)

Project Duration & Completion

Initial phase: 3 months sandbox access, AI Trailblazers 2.0: Up to 10 weeks, Successor AI CTO: Longer-term scaling support

Best Practices for a Successful Application

Start preparing use case ideas early - identify real problems where generative AI could add value

Ensure data readiness - structured data, text corpora, datasets, quality, labels, privacy/consent issues

Build a small AI/data team internally - having champions and engineers helps leverage sandbox fully

Engage early with Google/government/ecosystem partners for guidance and best practices

Plan for scaling post-prototype - think about transition from sandbox to production costs and infrastructure

Be realistic in MVP scope - choose focused prototype with measurable impact rather than broad ambitions

Attend workshops actively - maximise learning, networking, and insights from hands-on sessions

Document use of sandbox - model versions, infrastructure usage, experiment logs for evaluation

Adopt responsible AI practices from start - fairness, bias, privacy, explainability, governance

Monitor successor programmes like AI CTO/AI Cloud Takeoff for scaling opportunities

Sample Use Cases & Scenarios

GSK: GMP Assistant for internal document retrieval (saved ~450 man-days) and report generation tool (saved ~5,000 man-days/year)

AI Palette: AI tools for market research, trend prediction, and concept generation, dramatically cutting research time

Ethlas: Generative AI prototypes developed in sandbox environment

Document summarization, report generation, conversational interfaces, trend analysis applications

Real-world generative AI use cases across various industries and functions

Additional Information

Programme Evolution and Successor Initiatives: AI Trailblazers 2.0 Expansion: • Aims to empower up to 150 more organisations • Improved Innovation Sandboxes over up to 10 weeks • Access to Gemini (Google's newer foundation model) • Access to Duet AI for Developers for building, deploying, and operating applications AI Cloud Takeoff (AI CTO) - Successor Programme (2025): • New programme under Enterprise Compute Initiative (ECI) • Replaces or supplements AI Trailblazers programme • Helps companies scale AI adoption beyond prototyping • Incentives up to S$500,000 including consulting and cloud costs support • 70% grant for consulting, training credits, and additional cloud credits Key Benefits and Outcomes: • Access to advanced AI infrastructure at low risk during prototyping • Accelerated prototyping and iteration in sandbox environment • Hands-on learning and capacity building via workshops and mentoring • Validation via expert evaluation panels • Case studies and ecosystem momentum building Important Limitations: • Time-limited sandbox access (not full production support) • Competition/selection risk for cohort acceptance • Need internal capacity and commitment for success • Transition to production requires significant investment planning • Sustainability post-programme depends on organisation's ability to carry forward momentum