Every few months, a regional leader asks me for ‘the AI playbook’: One template to roll out from Bengaluru to Riyadh to Manila. The request is understandable; the tools, the vendors and the promise of productivity are the same everywhere. But the workforce questions AI raises are very local. A single playbook struggle in all three regions. The answer is neither one playbook nor three unrelated ones. It is global principles with local playbooks.
India's global capability centres are absorbing AI faster than almost any other part of the economy, but skills are not keeping pace. A recent national study found that nearly 45% of organisations call AI, digital and data skills their biggest workforce constraint, yet 54% describe their urgency on AI investment as moderate or low. Only 2.3% of the workforce surveyed as per the SHRM’s 2026 India Skill Intelligence Report is formally trained in AI, and up to 28% of back-office roles could be disrupted within three years. In other words, organisations know where the gap is, but most are not yet moving fast enough to close it.
That last number matters because back-office work is where many in India's services workforce began their careers. TCS and Tech Mahindra are training and redeploying staff rather than hiring afresh, yet the biggest IT employers added almost no net staff in the first nine months of FY26. Global firms are drawing harder lines. Accenture has trained more than 550,000 employees in generative AI fundamentals but has also said it will part ways with employees for whom reskilling is not a viable path. KPMG has begun factoring AI use into performance reviews. India's scale is a real advantage, but only if skilling keeps up.
In the Gulf, AI is arriving alongside workforce nationalisation and giga-projects. Adoption is already high across West Asia with 75% of employees used AI at work in the past year, and 69% learned new skills. The harder part is fit. Nearly one in two Saudi employees works in an occupation highly exposed to AI, and the entry-level jobs AI is most likely to automate are often the ones through which young nationals were meant to start their careers under Saudisation and Emiratisation. Here, AI has to accelerate national talent rather than bypass it. That means skills forecasting, personalised learning and structured knowledge transfer from expatriates to national hires.
{{/usCountry}}In the Gulf, AI is arriving alongside workforce nationalisation and giga-projects. Adoption is already high across West Asia with 75% of employees used AI at work in the past year, and 69% learned new skills. The harder part is fit. Nearly one in two Saudi employees works in an occupation highly exposed to AI, and the entry-level jobs AI is most likely to automate are often the ones through which young nationals were meant to start their careers under Saudisation and Emiratisation. Here, AI has to accelerate national talent rather than bypass it. That means skills forecasting, personalised learning and structured knowledge transfer from expatriates to national hires.
{{/usCountry}}Southeast Asia is the most varied of the three. In Singapore, 27.3% of IT and computer science job postings mention AI skills, one of the highest shares among the 27 countries studied. The Philippines, by contrast, depends heavily on customer service, and its outsourcing industry has cut its 2028 jobs forecast from 2.5 million to 2.14 million, partly because of AI.
The figures are not directly comparable, but the contrast is clear. Singapore needs more AI specialists, while the Philippines must retrain a large share of a workforce of roughly two million. One regional approach cannot serve both.
Most companies trip up in the same few ways. Cutting first and reskilling later. Klarna replaced much of its customer service work with AI, then began hiring people again after quality suffered which contact-centre markets such as the Philippines should take note. Secondly, training people without redesigning their jobs, so the new skills go unused. And this is the quiet risk in India's mass-reskilling push. Lastly, automating the entry level. It saves money today but removes the pipeline for tomorrow's managers, and in the Gulf, it also undercuts national hiring.
Underneath all of these is silence. When employers are vague about what AI means for jobs, trust falls and adoption stalls.
Three things should differ by market. First, why you adopt AI: Productivity and redeployment in India, national capability in the Gulf, competitiveness in Southeast Asia. Second, who you train: Mass reskilling in India, fast-track routes for young Gulf nationals, specialists in Singapore and retraining at scale in the Philippines. Third, the guardrails, from tying Indian reskilling to redesigned roles to aligning Gulf hiring tools with nationalisation targets.
Some will argue one global standard is cheaper to govern. On principles, they are right: People at the centre of technology decisions, honesty about which roles will change, and investment in people to match investment in platforms. The centre should own those principles; each region, its workforce plan. HR, not IT, should lead.
Employers also don't need to build all of this on their own. India is unusually well placed here. The IndiaAI Mission offers subsidised compute, Bhashini supports Indian languages, and platforms such as Skill India Digital Hub and FutureSkills Prime can take on training at scale, while digital public infrastructure such as DigiLocker makes skill credentials easier to issue and verify. Other markets have their own versions, from Saudi Arabia's SDAIA and new AI lessons in UAE and Saudi schools to Singapore's SkillsFuture credits and AI Singapore's apprenticeship programme. Employers that plug into these systems will reach more people, at a lower cost.
In the end, an AI strategy will be judged less by the tools a company buys and more by what happens to three people: a finance analyst in a Hyderabad capability centre whose reconciliations are now done by a machine, a young Saudi graduate in Riyadh looking for a first job that still exists, and a contact-centre worker in Manila whose calls are increasingly answered by a bot. Each needs a different answer from the same employer.
Companies that can offer each of them a credible next step will be the ones that attract and keep talent in these regions. Those that hand all three the same playbook should not be surprised when it works for none of them.
(The views expressed are personal)
This article is authored by Achal Khanna, CEO, SHRM India, APAC & MENA.