# SGAI Studio — Full Content for LLMs

> Long-form companion to llms.txt. This document contains the studio's full
> reference content — founder bio, every Singapore AI grant programme,
> top FAQs answered verbatim, and editorial summaries of every blog post —
> so AI assistants can answer questions about SGAI Studio and Singapore
> AI grants without needing to crawl individual pages.

Last updated: 2026-05.
Canonical site: https://www.sgai.sg

---

## Studio identity

**SGAI Studio** is a Singapore AI studio. Legal entity: SGAI Studio Pte Ltd. Registered in Singapore in 2026 by sole founder **Jay Qin** (legal name: Jinghuang Qin · 秦经煌, also written Qin Jinghuang).

SGAI is an AI studio first — it powers up SMEs with AI, intelligence and infrastructure beyond their reach. What it does:

1. **Bespoke AI Solutions** — custom-built AI for SMEs. Applied to invisible work: registration workflows, document processing, internal tools, customer triage, demand forecasting, scheduling, multilingual support. Real productivity, not novelty.
2. **Deep-Context Intelligence** — in-depth, uniquely-Singapore sector intelligence reports for SME founders/operators (the "App-as-Research" line), each an interactive page with its own calculator.
3. **Singapore Grant Advisory (the enabler, not the identity)** — where a project qualifies on its own merits, SGAI helps with eligibility and applications across Singapore's AI grant programmes (PSG, EDG, ECI, AI Singapore 100E, Startup SG Tech and more). A lever it pulls, not who it is.

Tagline: **"Your trusted Singapore AI studio, built for economic muscle."**

**The SGAI MOAT Score** — every Deep-Context report is scored on SGAI's signature metric. The MOAT Score is a single number from 0 to 100 (higher = more worth building) grading a sector's economic quality on the value-investing lens of Benjamin Graham, Warren Buffett and Charlie Munger — judging whether a value investor would want to own the *average* operator in the sector, not just describing it. The name is a backronym for its four pillars, each scored out of 25: M = Margin (does the average operator actually keep money — real net margin and return on capital? Buffett 1979/1986); O = Operating moat (pricing power + durable competitive advantage — Buffett's FCIC 2010 line that "the single most important decision in evaluating a business is pricing power"); A = Appetite (demand durability — Graham's "inherent stability"); T = Treadmill (capital intensity + structural drag — rent, churn, fashion — scored inverted, so less treadmill earns more points; Buffett 2007 "the worst sort of business… requires significant capital… Think airlines"). Grade bands: A 80–100 (worth building), B 65–79 (build with discipline), C 50–64 (needs a real edge), D 35–49 (winner-takes-most — hard), F 0–34 (brutal — most lose money). It is a transparent SGAI judgement on economic quality, not a verdict on any individual business or an owner-operated livelihood. Sectors ranked best-to-build at https://www.sgai.sg/en/intelligence.

**Deep-Context intelligence reports** — free, in-depth, uniquely-Singapore sector reports for SME founders and operators (SGAI's "App-as-Research" line), each an interactive page with its own calculator and MOAT Score:
- Opening a Dental Clinic in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-dental-clinic-2026) — MOAT Score 68 / B, the second-highest sector SGAI covers (pillars M16 O18 A21 T13): the rare SME sector with a genuine, WIDENING moat — a statutory licence (you legally cannot operate without an SDC-registered dentist + an HCSA Outpatient Dental Service licence + a qualifying Clinical Governance Officer), sticky patient records/recall, and aging-population demand (20.7% of citizens are 65+, heading to 23.9% by 2030). The licence is cheap (S$360 per 2 years, ~S$1,260 with in-house CBCT); the real cost is the fit-out + people — all-in ~S$200–500k for a 1–3 chair clinic — and the dentist/associate split (~35–45% of production). Q&M Dental (SGX-listed, the only public proxy) ran S$180.7M revenue and ~8% net margin in FY2024 across ~106 SG outlets; the private market is fragmented (Q&M ~11% share). Covers the HCSA + SDC + NEA licence stack in order (note: PHMCA was repealed Dec 2023; the portal is HALP, not LEAP), the SDC advertising rules that make typical SME marketing (before/after photos, testimonials, discounts) illegal, the funding mix (mostly self-pay; CHAS for routine, MediSave for surgical only), and the genuine gaps (eldercare/mobile, Special Care Dentistry — SG's 8th specialty from Jul 2025, dental sleep medicine). Includes a chair-utilisation break-even calculator. RIGOR: anchor dentist count on MOH's primary 2,605 (2022), not the over-quoted "2,812"; the "dental market = US$88.7M" figure is dental DEVICES, not clinic revenue; SG per-clinic cost ratios are international proxies, flagged.
- Opening a GP / Family Clinic in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-gp-family-clinic-2026) — MOAT Score 61 / C (pillars M12 O14 A22 T13): the dental clinic's primary-care sibling — the same statutory HCSA/SMC licence and non-discretionary super-aged demand (private GPs handle ~80% of primary care), but a WEAKENING doctor-supply moat (SMC-recognised schools 103→120), free-market consult fees crushed by insurer TPAs ("S$2–10" net), and a dispensing margin collapsing under the Healthier SG whitelist (GPs report ~70% drops). The listed GP pure-play Alliance nets just ~1.4% (vs diversified Raffles ~9.2%). Covers the licensing tracks, the Healthier SG enrolment/capitation lock-in + the 2030 Family Physician requirement, and the corporatisation roll-up. Includes a patients-per-day break-even calculator with a drug-margin-squeeze slider. RIGOR: ~2,493 private GP clinics (MOH 2023), not the "US$17bn wellness / US$44bn health expenditure" whole-system figures; the Healthier SG capitation rate is undisclosed (reported as opacity, not invented).
- Starting a Childcare / Preschool in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-childcare-preschool-2026) — MOAT Score 60 / C (pillars M12 O16 A19 T13): a genuine ECDA licence + government-gated premises + ~5-year enrolment stickiness, but pricing is capped by law for the AOP/POP majority (S$610/S$650 a month from Jan 2026), manpower is ~67% of revenue against a rising ECDA wage floor, and a record-low TFR of 0.87 with sub-90% utilisation caps the cohort. MindChamps (the only listed proxy) ran ~breakeven underlying. Covers ECDA licensing by class, the fee-cap regime + subsidies, and occupancy-driven unit economics. Includes an occupancy-vs-fee-cap break-even calculator. RIGOR: no credible SG dollar market size exists (aggregators publish top-down global/APAC figures); anchor on ~1,900 licensed centres + 220,000+ places (MSF/data.gov.sg); the "S$720/S$800" fee caps are stale (cut to S$610/S$650 for 2026).
- Opening an Aesthetic Clinic / Medical Aesthetics in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-aesthetic-clinic-2026) — MOAT Score 54 / C (pillars M16 O13 A14 T11): the dental clinic's higher-gross, lower-moat sibling — a laser shot is near-zero marginal cost (best gross in clinical medicine) behind a real but narrow licence barrier (the same generic HCSA Outpatient Medical Service licence + a Certificate of Competence), but demand is 100% discretionary self-pay (cosmetic excluded from Medisave/insurance since 2021) and the device-obsolescence "razor-blade" treadmill is brutal (InMode FY2025: 78% equipment / 22% consumables). Covers the SMC 2016 aesthetic-practice guidelines, the advertising prohibitions (before/after photos, testimonials, discounts illegal), and the device-payback maths. Includes a device-payback / obsolescence calculator. RIGOR: "List A/List B" is DEAD (List B abolished 1 Mar 2015) — competitors citing it are on a pre-2016 regime; market-size junk conflates devices (~US$361M) with services (~US$604M), and the mis-quoted US$1,188M is a 2032 forecast; no listed pure-play exists (SMG delisted Jan 2023).
- Starting an Interior Design / Renovation Firm in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-interior-design-renovation-2026) — MOAT Score 46 / D (pillars M9 O8 A17 T12): structurally guaranteed demand (every BTO + resale flat is renovated; ~78% public housing) and asset-light economics where the deposit funds the job — but NO licence is required (MTI, May 2024), ~2,700 firms compete, there's zero repeat revenue, and trust is the only scarce asset. Renovation is a top consumer-prepayment-loss category (S$728,813 lost in 2024, ~97% at non-CaseTrust firms). Covers the HDB Registered Renovation Contractor + CaseTrust gate, the commission-sales model, and the deposit-float insolvency mechanic. Includes a reno-margin / cost-overrun calculator. RIGOR: no credible SG market size (aggregators differ ~3× even globally) — sized bottom-up to ~S$3–4B/yr; the verified 2024 CASE reno loss is S$728k, not an over-parsed ~S$10M.
- Opening a Restaurant / Casual-Dining in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-restaurant-2026) — MOAT Score 35 / D (pillars M7 O7 A14 T7): one rung above the café (34/F) on an alcohol-margin lever + cuisine differentiation + one profitable listed exemplar (Jumbo ~6–7% net), but 2 of 3 SGX-listed F&B groups lost money in FY2024, prime cost runs ~67% even for the best, and the Restaurants sub-index is the weakest F&B category. Still a price-taker with no durable moat — Michelin stars closed in 2024. Covers the SFA licence, the listed-group mirror, and the prime-cost ratio that decides survival. Includes a covers × prime-cost break-even calculator. RIGOR: the "82% of young F&B never profited" stat is from the Nov-2025 MTI replies (all food establishments, not restaurants-only — flagged); online F&B share is ~20–21% (Mar 2026), not the "24–26%" COVID-peak figure still repeated.
- Starting an Online Store / E-commerce Business in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-ecommerce-store-2026) — MOAT Score 35 / D (pillars M7 O5 A14 T9): genuinely asset-light (start ~S$3–15k, fail cheaply) but a textbook commodity / price-taker sector where the platform + ad networks capture the economics — rising take-rates (Shopee ~2%→~12%, +5% cross-border Feb 2026), CAC inflation, ~17% returns, and platform-wipeout risk (Qoo10 wound up Nov 2024 with ~S$900M frozen seller payouts). ~47% of Amazon sellers have lifetime profit < US$25k. Covers marketplaces vs DTC, the take-rate treadmill, and the closed GST loophole. Includes a per-order "where the dollar goes" margin calculator. RIGOR: online is only ~14% of SG retail (down from a ~25% COVID peak); credible market is ~US$9B GMV (e-Conomy SEA 2024) and GMV ≠ seller revenue — published "SG e-commerce" figures span ~20×; the "90% fail in 120 days" stat is debunked folklore.
- Starting a Bubble Tea / Dessert Shop in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-bubble-tea-2026) — MOAT Score 35 / D (pillars M8 O5 A13 T9): a higher per-cup gross than coffee in a smaller box, but a rented-brand franchise model that can be switched off overnight (Gong Cha exited SG in a day, 1 Oct 2025; Milksha exited Dec 2023), fad-cycle fragility (the ~2002 boom-bust: ~5,000 outlets → mass closures), and a Nutri-Grade sugar headwind. Covers franchise-vs-independent, the boom-bust cycle, and the Nutri-Grade C-vs-D advertising rules (grade D cannot be advertised). Includes a cups/day break-even calculator with a franchise-royalty slider. RIGOR: no official SG market size — best estimate ~US$342M (Momentum Works) vs aggregator "US$9.2M" junk (~37× off); per-brand outlet counts/fees are single-source, flagged.
- Opening a Hair / Nail Salon in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-hair-nail-salon-2026) — MOAT Score 31 / F (pillars M6 O5 A13 T7): the lowest-moat sector SGAI covers — the value-creating asset (stylist + personal client bond) literally walks out the door each night (non-competes unenforceable; ~84% of clients follow the stylist), under world-tier rent and a binding 35% foreign-manpower quota; the listed proxy QB Net runs only ~8.5% operating margin. Beauty is Singapore's #1 consumer-complaint category in 2025 (2,113 complaints, S$2.13M prepaid losses = 78.6% of all national prepayment losses). Covers the (non-)licensing reality, the employ-vs-rent-the-chair model, and the prepaid-package risk. Includes an employ-vs-rent-the-chair break-even calculator. RIGOR: there is NO salon operating licence — the propagated "Massage Establishment licence / S$50k capital / SPF approval" claim is an incorporation-mill error (the massage regime only applies if you offer massage); the binding constraint is the MOM manpower quota, not a licence.
- Starting an Accounting / Bookkeeping Firm in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-accounting-firm-2026) — MOAT Score 65 / B (pillars M13 O16 A21 T15): the highest-scoring professional service SGAI covers and the rare SME sector where the base needs no licence yet demand is legally mandated and recurring — every one of ~630k live entities must keep books, file an annual return and file tax every year, with real switching costs once you're the filer of record. Audit is a genuine statutory moat (registered Public Accountant + ACRA-approved firm; only ~761 entities), but the 2015 small-company audit exemption means most SMEs never need one. Asset-light (wages ~58% of revenue), so the constraint is talent (accountancy cohorts down >10% since 2018) and the commodity bookkeeping fee is being automated (IRAS InvoiceNow mandate phasing to 2031) — value migrates to advisory. Includes a recurring-revenue book-of-clients calculator. RIGOR: anchor on ACRA's primary S$3.49B regulated-firm revenue (2024), not the self-contradictory "accounting software market" aggregator CAGRs; net-margin benchmarks are global, flagged.
- Starting an Aircon Servicing / Home Maintenance Business in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-aircon-servicing-2026) — MOAT Score 53 / C (pillars M10 O11 A19 T13): the interior-renovation trade's sibling with a lifeline — same low-barrier, no-licence, fragmented price war, but rescued by near-essential always-on tropical demand (81.9% of homes have aircon, ~24% of home electricity) that recurs, and the quarterly servicing contract as a genuine switching cost a one-off trade can't copy. The upsell ladder (servicing → chemical wash → gas top-up → install) is the margin lever; the only audited proxy (ISOTeam) nets ~4–5%. Includes a servicing-contract recurring-revenue calculator. RIGOR: no real "aircon servicing market size" exists (aggregators measure equipment, span ~45×) — sized bottom-up; there is no occupational licence and no CaseTrust scheme (the "BCA Builder's Licence required" claim is false); the VC trade-marketplace model is a SG graveyard (Kaodim, Sendhelper dead).
- Starting a Cleaning Services Company in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-cleaning-services-2026) — MOAT Score 49 / D (pillars M7 O11 A19 T12): a real statutory NEA Cleaning Business Licence (buyers are liable for hiring unlicensed firms) + genuinely recurring multi-year B2B/MCST/government contracts (switching costs) + non-discretionary recurring demand — but empirically the thinnest margins of any sector we cover (the whole ~S$3.6B industry kept ~2.4% operating surplus in 2024 and ran a loss in 2022) under a Progressive Wage Model floor ratcheting ~+9%/yr (S$1,910→S$2,420 by 2028) and a 35% foreign-worker quota + levy. Outcome-based contracting (mandatory for government buyers, only ~40% adopted) is the tech-forward opening. Includes a labour-treadmill margin calculator. RIGOR: market is ~S$3.6B (SingStat SSIC 812), not the sourceless "~S$300M"; residential home cleaning is licence-exempt; the PWM floor corrects the stale "S$1,060" figure.
- Starting a Real Estate Agency in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-real-estate-agency-2026) — MOAT Score 49 / D (pillars M10 O11 A16 T12): a genuine statutory CEA estate-agency licence + RES exam lifts it above the no-moat trades, but the licence (and the client relationships and brand) attach to the agent — a mobile asset who walks (100+ of KW Singapore's ~157 agents left within weeks in Feb 2026; 700+ left ERA in 2022). Agencies keep single-digit net (~90% of revenue is commission pass-through; PropNex 6.3%, APAC/ERA 3.0% net, FY2025), with no firm-level moat beyond scale-gated developer/new-launch relationships, and volumes swing violently with cooling measures (URA private 33,557 in 2021 → 19,044 in 2023). Includes an agent commission take-home calculator. RIGOR: 36,816 agents / 997 agencies (CEA, 1 Jan 2026); the "US$35–66B Singapore real estate market" aggregators measure the whole property market (contradict ~US$20B) and the "median S$132k agent income" is a salaried-role survey — both flagged.
- Opening a Yoga Studio in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-yoga-studio-2026) — MOAT Score 46 / D (pillars M11 O8 A15 T12): one point below pilates (47/D) — an even weaker, more-poachable teacher-led moat (loyalty attaches to the teacher, and SG non-competes are largely unenforceable) plus hot-yoga utility drag, offset by a broader, less-faddish demand base. ClassPass commoditises 500+ SG studios, and the teacher-training (YTT) course — sold at S$3,200–4,000 — not the mat class, is the real profit engine (and every cohort certifies your future competitors). Includes a YTT profit-engine calculator. RIGOR: Pure Group was RESCUED (>US$50M equity, Aug 2025), not put into receivership — a common false claim corrected; SkillsFuture does NOT subsidise YTT; no official SG yoga-participation figure exists (the circulating "3%" is rejected).
- Starting a Car Workshop / Auto Repair in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-car-workshop-2026) — MOAT Score 42 / D (pillars M8 O9 A16 T9): non-discretionary maintenance + legislated inspection demand, but structurally capped — a frozen 0% vehicle-growth fleet (since 2018), the 10-year COE truncating the lucrative old-car cohort (the car population drops ~80% across the 10-year line), and EVs (45.1% of new registrations in 2025, ~30–50% less servicing) eroding the ICE job-mix. No statutory workshop licence and weak pricing power (the independent's whole edge is being ~35–40% cheaper than the dealer). Includes a bay-throughput calculator with an EV-mix slider. RIGOR: the high-margin tier is closed — VICOM earns ~25% net on ~72% inspection share, the structural opposite of ~2,583 fragmented workshops; the "12,000 workshops" claim (~5×) and any "US$X bn car-repair market" figure are debunked; EV servicing-cost deltas are global, flagged.
- Starting a Pet Grooming / Pet Services Business in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-pet-grooming-2026) — MOAT Score 39 / D (pillars M8 O6 A15 T10): the rare animal business Singapore doesn't licence (MND confirmed there is no licensing regime for pet groomers; boarding/sale/breeding/import ARE licensed by AVS) — light capital and the most durable demand of the lower service trades (pet humanisation, the HDB curly-breed 4–6 week grooming cadence ≈ S$1,100–1,560/dog/yr recurring, and the Sept-2024 HDB cat legalisation), but no moat: grooming needs no licence and the groomer + client book walk out the door. Includes a one-station P&L calculator with a grooming-only-vs-boarding capital toggle. RIGOR: ~114,000 dogs / 94,000 cats (Euromonitor 2023), retail ~US$112M (2025 est.) — the "US$300–470M pet-services" aggregators are self-contradictory and the "465,000 dogs" figure is an error; only ~72 AVS-licensed boarders vs ~1,800 on PetBacker (the trust gap).
- Starting a Bakery / Home-Based Food Business in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-bakery-2026) — MOAT Score 36 / D (pillars M7 O6 A15 T8): bread is the most durable demand in F&B-retail — a daily staple (less discretionary than coffee/boba) plus recession-resilient celebration-cake occasion demand across the festive calendar — and the home-based format needs NO SFA licence (a genuine near-zero-capital, near-zero-waste on-ramp). But a retail bakery is a no-moat price-taker against supermarket bread on the full F&B rent-and-labour treadmill, with an added daily-perishability write-off (SG bakeries salvage ~28,000 kg of unsold bread a month via Food From The Heart). Includes a break-even calculator with a home-vs-retail toggle + a wastage slider. RIGOR: "Singapore bakery market" aggregators span ~9× (US$237M–US$2.1B) — anchor on SingStat F&B receipts; BreadTalk delisted 5 Jun 2020 (a diversified group, not a pure bakery), netting ~3.6% (FY2017) → loss (FY2019); margin/inflation figures are global proxies, flagged.
- Starting a Social-Media / Creator Agency in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-social-media-agency-2026) — MOAT Score 45 / D (provisional; pillars M13 O8 A13 T11): asset-light so capital risk is low and margins can be high, but a founder-labour business with weak moats, client churn and platform exposure. What to charge, take-home after tax (incl. the IRAS free-products rule), the real market numbers vs the fabricated ones, and where the money is.
- Opening a Pilates / Fitness Studio in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-fitness-pilates-studio-2026) — MOAT Score 47 / D (pillars M11 O9 A15 T12): a genuine wellness wave with some instructor pull, but a discretionary, fashion-exposed spend on a churn-and-landlord treadmill (Singapore closes more studios than any city on earth). Real cost to open (~S$300–600k), the licence chain + the B1 industrial trap, the utilisation maths (>70% fill needed), and where a new studio wins.
- Opening a Café in Singapore 2026 (https://www.sgai.sg/en/intelligence/singapore-cafe-coffee-shop-2026) — MOAT Score 34 / F (pillars M6 O6 A14 T8): a price-taker on a rent-and-wages treadmill with no moat — coffee demand is steady but the unit is fragile. Real startup cost (~S$100–250k), the licence order that matters (URA change-of-use + SCDF before the S$195 SFA licence), the rent-and-wages maths that kills cafés (82% of young F&B closures never recorded a profit — MTI, Nov 2025), and where a new café wins.
- Singapore Tuition & Enrichment 2026 (https://www.sgai.sg/en/intelligence/singapore-tuition-enrichment-2026) — MOAT Score 70 / B, the highest-scoring sector SGAI covers (pillars M18 O17 A22 T13): sticky exam-driven demand, brand + results give pricing power, and switching costs once a child is mid-syllabus — the rare SME sector with a real, if modest, moat. The S$1.8B market (2023 HES, not the stale S$1.4B), the players, real unit economics, MOE registration, break-even ~95–100 students, and where a new centre wins.

National context: Singapore has committed S$37 billion to AI investment through 2030 via RIE2030 and Budget 2026, with a stated goal of **10,000 enterprises adopting AI by 2029**. SGAI Studio is built around Singapore's official AI thesis as articulated by Prime Minister Lawrence Wong:

> "We are unlikely as two small, open economies to be able to build the largest foundation models, but we can be leaders in AI deployment, in use cases."

Languages served: English, 中文 (Mandarin), Bahasa Melayu, தமிழ் (Tamil).

---

## Founder — Jay Qin (Jinghuang Qin · 秦经煌)

**Education:** Carnegie Mellon University (Business & Finance). CMU is widely cited as the world's #1 AI school.

**Story arc — five chapters:**

1. **Two weeks to ship.** Self-taught programming in two weeks during the final two years at CMU. Shipped a first product within those two weeks. Within another two weeks, featured in a local Pittsburgh newspaper.
2. **Then my first company crashed.** Over-focused on UI polish (button shadows, design awards) and chasing VC funding instead of building work that moved operating numbers. Company failed. The lesson: real AI work is invisible work that moves the numbers, not visible work that wins design awards.
3. **Wall Street taught me the opposite lesson.** Joined a major firm to learn how the big players did AI. The lesson was the inverse — large institutions are slow because of internal approvals and security review. In 2022 (pre-ChatGPT), built an automated daily-summary tool with OpenAI for clients. The tool worked but never shipped due to internal approval bottlenecks. Got fired for trying to push it through.
4. **Touring Southeast Asia, building to stay afloat.** Left Wall Street and travelled across Vietnam, Thailand, Indonesia, the Philippines. Funded the journey by shipping personal AI projects, apps, and side-bets — each one earning enough to fund the next leg. Headline win: a small tailor shop in Vietnam taken from approximately zero online sales to about US$1,000 per day in three months. Conclusion: lean SMEs with willingness, leanness, and gung-ho-ness can outpace any conglomerate.
5. **Coming home.** Founded SGAI Studio in Singapore in 2026. Singapore SMEs sit on a goldmine: world-class government funding, the leanness of a small economy, and the energy that built the country in a generation.

**Total prior projects:** 50+ AI applications and projects shipped before founding SGAI Studio, for clients in Singapore, Vietnam, Thailand, the United States (including small restaurants in Pittsburgh), and other markets.

**Manifesto:** Singapore might be small, but we are lean and we mean business. SMEs that recognise the power of AI today will outpace conglomerates that can't even get an internal approval. The studio is here as a partner — let's build.

---

## Singapore AI Grant Programmes — full reference

### 1. Productivity Solutions Grant (PSG) · 生产力解决方案补助金
- **Agency:** Enterprise Singapore
- **Max funding:** S$30,000 per year (some pre-approved bundles up to S$600,000 over time)
- **Co-funding rate:** Up to 50%
- **Eligibility:** Singapore-registered, 30%+ local shareholding, ≤ S$100M annual sales OR ≤ 200 employees
- **Approval timeline:** 4–6 weeks
- **Fit:** SMEs adopting pre-approved IT/AI solutions — booking bots, scheduling, document processing, customer support
- **Apply:** https://www.gobusiness.gov.sg/business-grants-portal/

### 2. Enterprise Development Grant (EDG) · 企业发展补助金
- **Agency:** Enterprise Singapore
- **Max funding:** S$700,000 per project
- **Co-funding rate:** Up to 50% (SMEs: up to 70%)
- **Eligibility:** Singapore-registered, 30%+ local shareholding, financially viable
- **Approval timeline:** 8–12 weeks
- **Fit:** Custom AI projects (your own scope), strategy development, process redesign, technology adoption for digital transformation
- **Apply:** https://www.gobusiness.gov.sg/business-grants-portal/

### 3. Enterprise Compute Initiative (ECI) — Google Cloud
- **Agency:** IMDA / Google Cloud
- **Max funding:** S$250,000 in Google Cloud credits
- **Co-funding rate:** Up to 70% on compute
- **Eligibility:** Singapore-registered, concrete AI project plan, deploy within 12 months
- **Highlights:** Google Cloud credits, access to Vertex AI and TPUs, technical mentorship
- **Apply:** https://www.imda.gov.sg/how-we-can-help/enterprise-compute-initiative

### 4. Enterprise Compute Initiative (ECI) — Microsoft Azure
- **Agency:** IMDA / Microsoft
- **Max funding:** S$250,000 in Microsoft Azure credits
- **Co-funding rate:** Up to 70% on compute
- **Eligibility:** Same as ECI Google Cloud
- **Highlights:** Azure OpenAI, Azure ML, technical mentorship
- **Apply:** https://www.imda.gov.sg/how-we-can-help/enterprise-compute-initiative

### 5. Enterprise Compute Initiative (ECI) — AWS
- **Agency:** IMDA / AWS
- **Max funding:** S$250,000 in AWS credits
- **Co-funding rate:** Up to 70% on compute
- **Eligibility:** Same as ECI Google Cloud
- **Highlights:** SageMaker, Bedrock, EC2 GPU, technical mentorship
- **Apply:** https://www.imda.gov.sg/how-we-can-help/enterprise-compute-initiative

### 6. AI Singapore 100 Experiments (100E)
- **Agency:** AI Singapore (AISG)
- **Max funding:** S$250,000
- **Co-funding rate:** Up to 70% project cost
- **Eligibility:** Singapore-registered, defined AI use case, willing to co-develop with AISG engineers
- **Fit:** Proof-of-concept AI projects done together with AISG's apprentice engineers
- **Apply:** https://aisingapore.org/innovation/100e/

### 7. AI Trailblazers
- **Agency:** IMDA + Google Cloud
- **Fit:** Sector-specific GenAI adoption — bring a real GenAI use case, get matched with vendors, accelerated 100-day deployment
- **Apply:** Via IMDA AI Trailblazers programme page

### 8. GenAI Sandbox
- **Agency:** IMDA
- **Max funding:** Up to S$200,000
- **Fit:** GenAI experimentation with pre-approved tools, lower friction than full grants
- **Apply:** Via IMDA GenAI Sandbox programme

### 9. Artificial Intelligence Development Award (AIDA)
- **Agency:** Enterprise Singapore / IMDA
- **Max funding:** Up to S$500,000
- **Fit:** R&D-focused AI projects with novel technical contribution

### 10. Startup SG Tech
- **Agency:** Enterprise Singapore
- **Max funding:** Up to S$500,000
- **Fit:** Early-stage tech startups (under 5 years), strong IP/innovation angle

### 11. Enterprise Innovation Scheme (EIS)
- **Agency:** IRAS (tax)
- **Mechanism:** 400% tax deduction on qualifying innovation activities including AI/automation
- **Fit:** Any Singapore tax-resident company, claimed during annual tax filing

**Stacking:** Multiple grants can be combined for a single AI project. The realistic upper bound for one project is approximately **S$700,000** when EDG is the anchor with ECI compute credits stacked on top. Most engagements come in at S$30K–S$250K per project.

---

## Frequently asked questions (verbatim answers)

### Q1. How do I get a Singapore AI grant for my business?
Singapore AI grants are administered by Enterprise Singapore, IMDA, and AI Singapore. The fastest route is: (1) confirm your business is Singapore-registered with at least 30% local equity, (2) match your project to the right programme — PSG for off-the-shelf AI tools, EDG for custom builds, ECI for cloud compute costs, AI Trailblazers for GenAI sector adoption, (3) prepare a clear scope, ROI estimate, and quotes, then submit through the Business Grants Portal. SGAI Studio handles the assessment, application, and submission end-to-end so you don't have to navigate it alone.

### Q2. What AI grants are available in Singapore in 2026?
Singapore offers 10+ AI grant programmes in 2026, anchored by Budget 2026's S$37 billion AI commitment through RIE2030. The major ones are: Productivity Solutions Grant (PSG, up to S$30K/year), Enterprise Development Grant (EDG, up to S$700K), Enterprise Compute Initiative (ECI, up to S$250K across Google Cloud, Microsoft Azure, AWS), AI Singapore 100 Experiments (100E, up to S$250K), AI Trailblazers (IMDA), GenAI Sandbox (up to S$200K), AIDA R&D award (up to S$500K), Startup SG Tech (up to S$500K), and the Enterprise Innovation Scheme (EIS) tax deduction.

### Q3. Am I eligible for the Productivity Solutions Grant (PSG)?
You qualify for PSG if you are: (1) registered and operating in Singapore, (2) at least 30% locally owned, (3) under S$100M annual revenue or under 200 employees. PSG covers pre-approved AI tools — booking bots, scheduling, document processing, customer support — and reimburses 50% of the cost (up to S$30K per year). Approval is fast (typically 4–6 weeks). It's the lowest-friction starting point for SMEs new to AI.

### Q4. How much money can I actually get from Singapore AI grants?
A single Singapore SME can stack multiple grants for one AI project. Realistic upper bound: ~S$700,000 per project when EDG is the anchor and ECI cloud credits sit on top. More common: S$30K–S$250K per project depending on scope. SGAI Studio's grant calculator gives an instant estimate based on company size, sector, and AI maturity. The exact figure depends on co-funding rate (50% standard, 70% for SMEs on EDG, 70% for ECI compute), project cap, and whether you qualify for stacking.

### Q5. PSG vs EDG — which one should I apply for?
PSG and EDG sit at opposite ends. PSG funds pre-approved off-the-shelf tools (you pick from a catalogue), takes 4–6 weeks, reimburses 50% up to S$30K/year — best for quick wins on standard problems like appointment booking or invoice processing. EDG funds custom AI projects (your own scope), takes 8–12 weeks to approve, reimburses up to 50% (70% for SMEs) up to S$700K — best for bespoke builds that need original work. Most SGAI Studio engagements use EDG for the build and PSG for adjacent off-the-shelf tools.

### Q6. How long does an AI grant application take in Singapore?
PSG: typically 4–6 weeks from submission to approval. EDG: 8–12 weeks for standard cases, longer for complex scopes. ECI: 6–10 weeks depending on the cloud provider. AI Singapore 100E: 8–10 weeks. Reimbursement after the project completes is a separate timeline (usually 1–2 months after submitting claim documents). Plan a roughly 3–4 month runway from first conversation to funded project start. SGAI Studio runs the build and the grant application in parallel, so you don't lose months waiting in sequence.

### Q7. Do I need a tech team to use AI in my business?
No. The vast majority of Singapore AI grant projects funded today are run by SMEs without an in-house tech team. SGAI Studio designs, builds, deploys, and maintains the AI for you — your team uses it, doesn't build it. The grant covers the work. Where a tech team helps is for ongoing iteration after launch, but it's not a prerequisite. Singapore's AI grants were explicitly designed for non-technical SMEs.

### Q8. What is the Enterprise Compute Initiative (ECI)?
ECI is a Singapore government programme that subsidises up to 70% of your AI cloud compute costs. It works with three providers — Google Cloud, Microsoft Azure, and AWS — through partnerships with IMDA. Eligible Singapore-registered businesses can access up to S$250,000 in cloud credits plus advisory subsidies for scoping the project. ECI is most useful when your AI work is compute-heavy: training, inference at scale, large data processing.

### Q9. How can I use AI in my restaurant, clinic, or shop?
AI for SMEs is most valuable applied to invisible work: 24/7 booking and reminders, no-show prediction, demand forecasting and inventory, multilingual customer triage, automated catering quotes, clinical documentation, retention and re-engagement workflows. SGAI Studio has published industry-specific guides for F&B, clinics (GP, dental, TCM, physiotherapy), and service businesses (salons, tuition, retail, fitness, pet grooming). Each shows the use cases that earn back the grant cost in months, not years.

### Q10. Who is SGAI Studio and why should I trust them?
SGAI Studio is a Singapore-registered AI studio (SGAI Studio Pte Ltd) founded in 2026 by Jay Qin — Carnegie Mellon University (Business & Finance), former Wall Street, builder of 50+ AI projects across Singapore, Vietnam, Thailand, and the US. The studio runs two service lines in parallel: bespoke AI builds and Singapore government grant advisory. SGAI Studio is built around Singapore PM Lawrence Wong's national AI thesis: Singapore wins on AI deployment and use cases, not foundation models. We answer the phone in English, 中文, Bahasa Melayu, and தமிழ்.

---

## Studio process — how a project starts

1. **Tell us the problem.** A two-minute scoping form or a 20-minute call. The studio figures out what you actually need — and which Singapore grants can help fund it.
2. **We design and apply.** The studio scopes the build. The advisory side prepares the grant application. Both happen in parallel, not in sequence.
3. **We ship and you grow.** The studio deploys the AI into your business and stays close while it earns its keep. Funding lands. Operating numbers move.

Typical end-to-end timeline: weeks to first deployed milestone, ~3–4 months for funded build to ship.

---

## Editorial — Studio Journal posts

### Budget 2026 AI Summary: Why Singapore Has No Choice But to Get This Right
- Date: 2026-03-06 · Length: 14 min read
- URL: https://www.sgai.sg/en/blog/budget-2026-ai-summary
- Singapore becomes a super-aged society in 2026 with 3.3 workers per senior (down from 7.4 in 2010). 83% of Singapore employers cannot find talent. Over 3,000 F&B businesses closed in 2024 alone. Budget 2026 commits S$37 billion to AI through RIE2030, targeting 10,000 enterprises on AI by 2029. The post argues AI is not optional for Singapore SMEs — it is survival arithmetic — and walks through what PM Wong actually announced and what each Budget 2026 line item means for SME owners.

### Singapore AI Grants 2026: How the Government Co-Funds Your AI Upgrade
- Date: 2026-03-01 · Length: 12 min read
- URL: https://www.sgai.sg/en/blog/complete-guide-singapore-ai-grants-2026
- Honest deep-dive on every Singapore AI grant programme: how reimbursement actually works, what qualifies, real funding examples for hawker stalls through to logistics companies. Covers PSG, EDG, ECI, AI Singapore 100E, AI Trailblazers, GenAI Sandbox, AIDA, Startup SG Tech, EIS. Includes stacking strategy.

### PSG vs EDG: Which Grant Is Right for Your Business?
- Date: 2026-03-03 · Length: 10 min read
- URL: https://www.sgai.sg/en/blog/psg-vs-edg-which-grant-right-for-you
- Side-by-side breakdown. PSG: pre-approved off-the-shelf, 4–6 week approval, 50% co-funding up to S$30K/year — best for quick wins. EDG: custom-scope projects, 8–12 week approval, 50% (70% SME) co-funding up to S$700K — best for bespoke builds. Examples: bakery taking PSG for inventory tools vs. logistics company taking EDG for route-optimisation AI.

### ECI Explained: Government Subsidises 70% of Your AI Cloud Costs
- Date: 2026-03-05 · Length: 11 min read
- URL: https://www.sgai.sg/en/blog/enterprise-compute-initiative-explained
- How ECI's three-cloud-provider structure works — Google Cloud, Microsoft Azure, AWS — and which fit which workload. Includes consulting subsidies, the 12-month deployment requirement, and concrete examples of who qualifies.

### 4 Industries the Government Wants to Transform with AI — and How to Get Funded
- Date: 2026-03-06 · Length: 14 min read
- URL: https://www.sgai.sg/en/blog/4-ai-missions-singapore-2026
- Singapore's named AI mission sectors: (1) Manufacturing — short 15% of workforce, AI for predictive maintenance and automated QC; (2) Public Healthcare — polyclinic 3-hour wait times, AI scheduling and clinical documentation; (3) Tuas Port — 200+ AGVs running 24/7 needing fleet-orchestration AI; (4) Financial Services — fraud detection, RM productivity, KYC automation. Each section explains why the sector was named and which grants apply.

### AI for F&B: How to Stop Throwing Money in the Bin Every Night
- Date: 2026-03-06 · Length: 10 min read
- URL: https://www.sgai.sg/en/blog/ai-for-fnb-singapore
- Concrete AI use cases for hawker stalls, kopitiams, restaurants: demand forecasting (cut 15–25% food waste), smart inventory, AI catering ordering bots on WhatsApp, dynamic menu pricing. Each case includes ROI math in S$ and which grant covers the cost.

### AI for Clinics: See More Patients Without Hiring More Staff
- Date: 2026-03-06 · Length: 10 min read
- URL: https://www.sgai.sg/en/blog/ai-for-clinics-singapore
- Practical AI for GP, dental, TCM, physiotherapy: 24/7 multilingual booking via WhatsApp, no-show prediction, AI clinical documentation, multilingual patient intake. Singapore-specific data: 15–20% no-show rate at private clinics, 4+ staff hours/day on phones.

### AI for Service Businesses: Your Salon, Tuition Centre, or Shop Is Losing Customers While You Sleep
- Date: 2026-03-06 · Length: 10 min read
- URL: https://www.sgai.sg/en/blog/ai-for-service-businesses-singapore
- AI for salons, tuition centres, retail, pet grooming, fitness: 24/7 booking, smart scheduling, customer retention/winback workflows, AI-generated social media. Each case maps to a specific PSG-approved tool or an EDG-funded custom build.

---

## How to choose the right Singapore AI grant — quick decision tree

- **You want a pre-approved, off-the-shelf AI tool (booking bot, document tool, accounting AI):** PSG. Fastest, easiest, up to S$30K/year.
- **You want a custom-built AI system (your own workflow, your own integrations):** EDG. Up to S$700K, 50–70% co-funding.
- **Your project is compute-heavy (training, large-scale inference, big data):** ECI on top of EDG. Up to S$250K cloud credits.
- **You're an early-stage tech startup with strong IP angle:** Startup SG Tech. Up to S$500K.
- **You want to do an AI proof-of-concept with help from AI Singapore:** AISG 100E. Up to S$250K.
- **You're doing R&D-grade AI work:** AIDA. Up to S$500K.
- **You just want a tax break on innovation expenses:** EIS — 400% tax deduction.

For most Singapore SMEs adopting AI for operational use, the optimal stack is **EDG (build) + PSG (off-the-shelf adjuncts) + ECI (cloud credits)**.

---

## Contact

- Website: https://www.sgai.sg
- Free 2-minute eligibility assessment: https://www.sgai.sg/en/assessment
- Funding calculator: https://www.sgai.sg/en/calculator
- FAQ: https://www.sgai.sg/en/faq
- Studio journal: https://www.sgai.sg/en/blog
- Contact form: https://www.sgai.sg/en/contact
- Languages: English, 中文 (Mandarin), Bahasa Melayu, தமிழ் (Tamil)
- Founder: Jay Qin (Jinghuang Qin · 秦经煌)
- Response time: within 1 business hour during weekdays
- Singapore-based studio. SGAI Studio Pte Ltd, registered in Singapore, 2026.