Issue 017 built the index, 018 printed the corridor's data, 019 watched the institutions answer, 020 followed the chain to where it breaks. 021 stands at the one door where more people are going in than coming out, and counts them separately.
Start with the number, and then immediately with its method, because printing one without the other is how this figure will be misused everywhere else. Nomura economists Sonal Varma and Si Ying Toh published a note on 7 August tallying AI-related hiring and AI-related job losses across Asia. India comes out positive: 83,100 AI-related hires against 31,921 AI-related layoffs and attrition since 2022, a net of about 51,000, and the only major market in their tally where the ledger runs that way. Across Asia the split is 130,758 created against 60,654 eliminated, with more than 90% of the creation in technology and about 60% of the elimination in financial services. Now the method: this is a count of 69 anecdotes gathered mostly between 2022 and August 2026, not a survey and not an official series, and Nomura says so, describing its own findings as indicative of broader trends rather than absolute job numbers. It is sell-side research. Treat the shape as real and the decimal places as decoration.
The spine of this issue is not that number. It is the sentence Nomura puts underneath it. Most firing cases, the note says, are support teams replaced by chatbots, while most hiring cases are of IT-services graduates that firms explicitly credit to AI demand. Read that twice, because it describes two entirely different populations moving through the same doorway in opposite directions, and the aggregate we just printed is the arithmetic of subtracting one from the other as though they were the same people. They are not. Nomura goes further than we would have dared to on its behalf: since displaced workers rarely transition to AI engineering roles, it writes, the aggregates do mask distributional pain. The bank that produced the good news also printed the caveat that dismantles the good news, and almost nobody quoting the +51,000 has carried the second half.
The supporting data says the same thing in a different register. Naukri's JobSpeak index for July, released on 3 August, put overall white-collar hiring up 5% year on year with the index at 3,227, and AI and machine-learning roles the strongest-performing segment at 33% growth. Underneath that headline sits the number that matters here: hiring for candidates with 13 to 16 years of experience rose 67% year on year, with the 16-plus band up 49% and the 8-to-12 band up 40%, while fresher hiring grew 6%. The AI premium is not distributed evenly across the entering queue. It concentrates hard up the experience curve. That is consistent with Indeed's mid-July India tracker, where 66% of employers reported paying salary premiums for AI-skilled staff while 54% of employees already working in AI-exposed roles said their pay had stayed flat or fallen. The premium goes to the person being hired, not to the person already there.
TCS is both queues inside one company, which is why it is worth stating carefully rather than dramatically. Its headcount closed the June quarter at 593,798, a net addition of 9,279 on the previous quarter and, on analyst arithmetic reported by Business Standard, the largest sequential addition in fifteen quarters. That follows a financial year in which the same company's headcount fell by 23,460 net. Management has described converting roughly 1% to 1.5% of the workforce into forward-deployed AI engineers, and reports an annualised AI revenue run rate of $2.6bn in the quarter. Note what is not established anywhere: that the people added are the people who left, or that any of them are. No Indian IT services firm publishes a redeployment rate. The one metric that would settle whether these are two queues or one is the metric nobody discloses.
India's net AI-related job creation since 2022 on Nomura's tally, 83,100 hires against 31,921 losses, the only major market where the number comes out positive. Print the method in the same breath or not at all: it is a count of 69 anecdotes gathered mostly between 2022 and August 2026, which Nomura itself calls indicative of broader trends rather than absolute job numbers, and it is sell-side research. The figure spans nearly four years, not this fortnight.
The reason to run this number is not that it is good news but that the bank which produced it printed the sentence that dismantles it: most firing cases are support teams replaced by chatbots, while most hiring cases are of IT-services graduates that firms explicitly credit to AI demand, and since displaced workers rarely transition to AI engineering roles, the aggregates do mask distributional pain. Two populations are moving in opposite directions through the same doorway, and a net figure is the arithmetic of subtracting one from the other as though they were the same people, which is the one thing the source says they are not.
Issue 017 built the index. Issue 018 printed the corridor's data. Issue 019 watched the institutions answer. Issue 020 followed the offshore chain to the point where it breaks and nobody files anything. This issue turns to the one major market where the chain is visibly running the other way, and asks the question the aggregate is designed not to answer: when a country's AI jobs ledger comes out positive, who exactly is in the plus column?
The West spent this fortnight on other things. Jeff Dean and three of Google's most decorated researchers left to found Discovery Loop. OpenAI, Meta and xAI shipped incremental updates. And on 2 August the EU AI Act reached a milestone that turned out to be two milestones running in opposite directions: Article 50's chatbot disclosure obligation came into application, so a European user must now be told when they are talking to a machine, while the high-risk obligations covering AI used in recruitment, candidate selection, performance evaluation, promotion and termination were pushed back to 2 December 2027. That delay is not a proposal. It is Regulation (EU) 2026/1744, the Digital Omnibus on AI, in force since 27 July. Europe now requires the chatbot to introduce itself and has given the hiring algorithm another sixteen months of quiet.
So the fortnight's actual distribution of attention looks like this. Brussels regulated the disclosure and deferred the employment protection. Silicon Valley conducted a leadership drama and shipped point releases. And a sell-side note that almost nobody outside Asia read established that in the largest market in the world for this kind of work, AI has so far created more jobs than it has destroyed, while quietly noting that the created and the destroyed are different people and nobody is tracking the gap between them. The story is not where the benchmark decimals moved. It is who is standing in which queue.
The Gulf runs the deliberate version of the same sorting, which is what makes it useful as a comparison rather than a distraction. On 9 August the National Committee for the Agentic AI Project convened more than 100 federal officials in the UAE and launched the project's strategic track across seven pillars, restating the target of converting 50% of federal operations, procedures and services to agentic AI within two years under the doctrine that human leads and AI enables. Underneath it, an April framework put ministers and directors general under performance assessment on the speed of their AI adoption, with 80,000 federal employees in training. In India the sorting between the two queues is being done by the labour market and recorded by nobody. In the UAE it is being done by the state, on a published timetable, with a named person accountable in each entity. Neither is obviously kinder. Only one of them is legible.
The composition is visible in the supporting data even though the transition is not: Naukri's July index, released 3 August, has AI and machine-learning hiring up 33% year on year, but within it demand for candidates with 13 to 16 years of experience rose 67% and the 16-plus band 49%, against 6% growth in fresher hiring, while Indeed's mid-July India tracker found 66% of employers paying salary premiums for AI-skilled staff and 54% of employees already in AI-exposed roles reporting pay flat or falling; the premium is concentrated on scarce senior experience being bought in, not on the workforce absorbing the technology where it already sits.
Put the strongest counterargument first, because it is genuinely strong. The number we are building on is a tally of 69 anecdotes assembled by a sell-side bank, not a labour-force survey. Nomura says plainly that it is indicative rather than absolute. The net figure spans nearly four years, not this quarter, so the fortnightly framing risks implying a velocity nobody claimed. Naukri's index is a job-postings measure from one portal. Indeed's India tracker is effectively a single press release reproduced by three trade outlets, and the underlying report is not published anywhere we can reach. On the sceptical reading, this issue is an argument about composition built on aggregates too soft to carry it.
The reply is that the softness of the aggregate is the argument, not a weakness in it. We are not asking anyone to believe 83,100 to the nearest hundred. We are pointing at the structure of the claim: that every number in this story, hard or soft, describes flows of jobs rather than flows of people, and that flow-of-jobs data cannot distinguish between an economy that redeployed its displaced support staff and an economy that replaced them with a different cohort entirely. Nomura, having done the tally, says explicitly that displaced workers rarely make that transition and that the aggregates therefore mask distributional pain. That is not our inference laid over their data. It is their sentence. And it is testable: any firm could publish a redeployment rate tomorrow and settle it. None does.
The Australian coda shows how quickly this becomes contested once you try to trace an individual person rather than a category. Commonwealth Bank confirmed cutting 176 technology and engineering roles, part of a wider 276 across the bank. The Finance Sector Union alleges that roles with similar titles and functions were subsequently advertised at CBA India. The bank denies it, stating that the roles currently advertised in India are separate and unrelated to the proposed changes in Australia. We cannot adjudicate that, and we are not going to pretend otherwise. What we can say is that the dispute is only possible because no disclosure regime requires anyone to answer it, and that the same opacity which makes the Australian claim unprovable makes the Indian aggregate unfalsifiable. One country's union suspects a transfer it cannot demonstrate. Another country's tally records an arrival it cannot attribute.
For anyone standing in either queue, the practical reading is that the aggregate will not protect you and was never about you. If you are in a support, service or process role that a chatbot can absorb, the fact that your national economy is net-adding AI jobs tells you nothing about your own prospects, because the roles being added ask for thirteen to sixteen years in a discipline you have not been working in. If you are being hired into that premium, understand what you are being paid for and how long the scarcity lasts, because a 67% surge in demand for your experience band is a supply signal that the market is already responding to. And if you run a workforce, the number worth publishing is the one nobody publishes: of the people whose roles you automated, how many are still with you, and doing what.
We are not claiming India's AI hiring boom is fake; the direction is real and the counter-programming value is real, because the West spent this fortnight on a leadership drama and a benchmark decimal while Brussels made the chatbot introduce itself and gave the hiring algorithm until December 2027. We are claiming something narrower: that a net figure cannot tell you whether an economy redeployed its displaced workers or replaced them, that the only metric which could is a redeployment rate, and that no firm in this story publishes one. India's ledger is the best news in AI and employment this year, and it is unfalsifiable in the exact place where it matters most.
The panel this issue is deliberately not a week-on-week comparison, and it is not a layoff split either. It is the two queues, printed side by side, from the same source and the same tally: 83,100 AI-related hires in India since 2022 against 31,921 AI-related layoffs and attrition over the same period. Do not read the difference as a net gain accruing to a population. Read the two figures as what they are, two separate flows of separate people through the same labour market, which Nomura counted separately and then subtracted because subtraction is what a ledger does. The subtraction is arithmetically fine and humanly meaningless, and the bank says as much when it notes that displaced workers rarely transition into the roles being created.
Every rising role this issue comes from the same absence. Somebody has to measure the thing between the two queues. A redeployment rate is not a complicated metric: of the roles you automated in a period, what share of the people holding them are still employed by you, and in what. It does not exist as a disclosure anywhere in the Indian IT services sector, or in the Australian banking sector, or in the Gulf public sector converting half its operations to agentic delivery. Until it does, every claim in this fortnight's story, ours included, is an argument about categories conducted by people who cannot see individuals.
The Gulf offers the one place where the sorting has an owner. The UAE's agentic project now runs a strategic track across seven pillars with ministers and directors general assessed on adoption speed and 80,000 federal employees in training, which means that somewhere in each federal entity there is a person whose performance review depends on converting services to agentic delivery. That is a real job with a real timetable, and it is the mirror image of the Indian situation, where the same transition is happening faster and at greater scale with nobody accountable for where the displaced went. Neither arrangement has published a redeployment rate. Only one of them has someone whose name is attached to the question.
The individual move is to stop reading national aggregates as personal forecasts. India's AI jobs ledger running positive is a true statement about India and a useless statement about any individual Indian worker, in exactly the way that a rising index tells a delisted shareholder nothing. What is actionable is the composition underneath: the premium concentrating in the 13-to-16-year band, the 6% growth in fresher intake against 33% growth in AI roles overall, and the 54% of people already in AI-exposed work whose pay has not moved. Those three figures describe a market paying hard for scarce experience, paying little for new entrants, and paying nothing extra to the people who absorbed the technology into jobs they already had. That last group is the largest and the least discussed.
6 rising role categories, each with a sourced hiring signal.
Nomura counted the hires and the losses separately and then said the displaced rarely move into the roles being created, which means the metric that would settle the question does not exist at any firm in this story. Building and publishing it is unclaimed work with an obvious first mover.
Named Nomura
TCS has described converting roughly 1% to 1.5% of a 593,798-strong workforce into this role while reporting an annualised AI revenue run rate of $2.6bn for the June quarter. It is the single most concretely specified destination role in the Indian market right now.
Named TCS
Cognizant expects its first Frontier cohort by the fourth quarter of 2026, against a stated ambition of 5,000 Frontier Certified Engineers and 10,000 Frontier Business Operators with no date attached. Someone has to design the path from the second population to the first.
Named Cognizant
Naukri's July index shows AI and machine-learning demand up 67% year on year in the 13-to-16-year experience band and 49% in the 16-plus band, against 6% for freshers. The scarce commodity is not AI skill, it is AI skill attached to a long career, and that is now a distinct hiring brief.
The UAE's April framework puts ministers and directors general under performance assessment on the speed of AI adoption, with 80,000 federal employees in training and a strategic track launched on 9 August. That makes converting services to agentic delivery a named accountability rather than a project.
Commonwealth Bank confirmed 176 technology and engineering cuts, the Finance Sector Union alleges similar roles were re-advertised in India, and the bank denies any connection. Nobody can adjudicate that today because no disclosure regime requires an answer, which is precisely the gap.
Named Commonwealth Bank
Printed, not charted. These figures are not measured the same way, on any of the four counts that would let them share a scale. Drawing them together would suggest a comparison the sources do not support, so the numbers are set out instead.
Queue out (India, since 2022)
Queue in (India, since 2022)
This week's signal through the India, Middle East and Africa lens.
ACCELERATING
Signal
India is the one major market where the AI jobs ledger runs positive, and the only one where the composition underneath it can be read. Nomura's 7 August note counts 83,100 AI-related hires since 2022 against 31,921 AI-related losses, within an Asia-wide tally of 130,758 created and 60,654 eliminated, more than 90% of the creation in technology and about 60% of the elimination in financial services. Naukri's July index, released 3 August, has overall white-collar hiring up 5% and AI and machine-learning roles up 33%, with the 13-to-16-year experience band up 67% and freshers up 6%. TCS closed the June quarter at 593,798 people, a net addition of 9,279 reported as the largest in fifteen quarters, after a financial year in which its headcount fell 23,460 net, and reports annualised AI revenue of $2.6bn. Cognizant expects its first Frontier cohort by the fourth quarter. On the demand side, foundit's tracker puts 64% of new global capability centre roles created in 2026 as requiring AI, data science or intelligent automation skills.
BUILDING
Signal
The Gulf is running the same sorting deliberately, with the state as employer. On 9 August the National Committee for the Agentic AI Project convened more than 100 federal officials and launched the project's strategic track across seven pillars, restating the target of converting 50% of federal operations, procedures and services to agentic AI within two years under the doctrine that human leads and AI enables, in what Huda Al Hashimi described as the operational start of deploying AI models. The machinery underneath is what makes it a workforce story: an April Cabinet framework places ministers and directors general under performance assessment on the speed of AI adoption, and 80,000 federal employees are in training. This is the civil servant with an AI key performance indicator, which is a different proposition from a technology programme. Saudi Arabia continues to build the physical layer, with DataVolt expecting to break ground within two to three months on a 1.5GW, $5bn AI campus at NEOM's Oxagon.
EMERGING
Signal
Last issue followed a Sydney automation decision to a call centre outside Johannesburg that nobody counted. The sequel is a dispute nobody can settle. Commonwealth Bank confirmed cutting 176 technology and engineering roles as part of a wider 276 across the bank; the Finance Sector Union alleges that roles with similar titles and functions were subsequently advertised at CBA India; and the bank states on the record that the roles currently advertised in India are separate and unrelated to the proposed changes in Australia. We cannot adjudicate it, and the reason we cannot is the same reason India's positive ledger is unfalsifiable: no disclosure regime anywhere requires an employer to say where automated work went. Meanwhile Kenya is buying into the category at the other end. Invest Kenya and the Outsourcing Alliance of Kenya signed a Cooperation Agreement on 4 August to position the country as Africa's premier global business services hub, one issue after we followed AI deleting exactly this kind of work.
Short read · this week's signal across the nine sectors we cover
Indian IT services is the entering queue in this issue: more than 90% of Asia's AI-related job creation on Nomura's tally sits in technology, and TCS added a net 9,279 people in the June quarter after shedding 23,460 net across the previous financial year, while converting 1% to 1.5% of its workforce into forward-deployed AI engineers.
Support and process delivery is the leaving queue. Nomura's characterisation is that most of the firing it counted is support teams replaced by chatbots, and the roles being created against them ask for thirteen to sixteen years in a different discipline, which is why the two flows do not net against each other in any way that helps an individual.
About 60% of Asia's AI-related job eliminations on Nomura's tally are in financial services, and the fortnight's clearest illustration is a bank: Commonwealth Bank confirmed 176 technology and engineering cuts within a wider 276, with a contested union allegation about Indian re-advertising that the bank denies.
The UAE is the one employer in this issue that has attached names and dates to the transition: a strategic track launched on 9 August across seven pillars, a 50% conversion target within two years, ministers and directors general assessed on adoption speed, and 80,000 federal employees in training.
Certification is where the two queues would have to meet if they ever did. Cognizant expects a first Frontier cohort by the fourth quarter of 2026 against a stated ambition of 5,000 Frontier Certified Engineers and 10,000 Frontier Business Operators, and 64% of new global capability centre roles created in 2026 are reported as requiring AI, data or automation skills.
Europe made the machine introduce itself. Article 50 of the AI Act applied from 2 August 2026, so a user must be told when they are interacting with an AI system, with machine-readable marking of synthetic content following for grandfathered systems from 2 December 2026.
Platform and quick-commerce work remains the contested edge of the Indian labour question, with the challenge to the Karnataka Platform Based Gig Workers Act consolidating rather than fragmenting after Uber's petition was tagged into the pending IAMAI batch on 28 July.
Compute capacity is the one part of the Gulf's AI build with a physical footprint and a date: DataVolt expects to break ground within two to three months on a 1.5GW, $5bn campus at NEOM's Oxagon, targeting operations from 2028.
Occupational and mental-health provision for AI supply-chain workers stays open from 019 and 020: Kenya's draft policy would make it a duty rather than a benefit, and nothing has been published on the consultation's outcome.
For Editor reAImagine · curated to this issue's signal · 90-day horizon
Why now
Nomura counted both queues and then said the displaced rarely cross between them, which makes the crossing rate the only decisive metric, and the person who builds it becomes the Redeployment rate owner.
Do this
Why now
A 33% rise in AI roles conceals 67% growth in the 13-to-16-year band against 6% for freshers, and the person who reads the bands rather than the headline becomes the Senior-band AI capability lead.
Do this
Why now
Certification programmes are scaling with no published date and no stated cohort size, and the person who specifies where trainees actually land becomes the Certification programme architect rather than its customer.
Do this
Why now
The UAE assesses ministers and directors general on adoption speed with 80,000 employees in training, and the person who can run that conversion end to end becomes the Agentic conversion officer.
Do this
Why now
A country's ledger running positive says nothing about any individual's queue, and the person who can trace an actual transfer rather than a category becomes the Offshore transfer investigator.
Do this
Nothing scores this issue, for the fourth consecutive time, and that is honest rather than convenient: no entry's resolve-by date has passed or falls in the coming week, with the nearest being LEDGER-001-05 on Accenture's Q4 results around late September and LEDGER-001-03 on 30 September. Two entries get real updates. LEDGER-002-02 closes a handover debt: every figure in Challenger's July report is now verified against the company's own publication, with 33,429 US cuts in July, 477,033 year to date, AI cited in 112,713 or about 24% of all cuts, and AI leading all July reasons at 10,970 or 33%, a fifth straight month. One caveat belongs in print: the first-half baseline of 101,743 is not stated in the July report. It is exactly implied by subtracting 10,970 from 112,713, and it is confirmed by the June report, but from the July document alone it is derived rather than quoted and we cite it that way. On the trajectory, one month of H2 at that rate leaves the second half short of the first, and the August report around the start of September is the second data point. LEDGER-001-03 gets a negative update, which is still an update: no qualifying announcement from any of the four in the fortnight to 13 August, and TCS's only release in the window was a single-vendor Gemini experience centre in Mexico, which is arguably movement in the opposite direction. It stays trending toward a hit on 020's printed reasoning and will be scored on 30 September. LEDGER-004-03 is unchanged and still heading for a miss, with nothing from MoHRE in the fortnight and the July compliance-positive posture standing. LEDGER-001-04 carries a criteria risk worth stating now rather than at resolution: every hit we can find is in the Chief AI Officer family, and if the sovereignty mandate is being absorbed into CAIO roles rather than spawning a distinct title, we will be scoring a definitional miss rather than a real-world one. Two corrections to our own register follow below, both from this issue's verification pass, and both concern claims we carried in 020's source register rather than in its screens. Ledger 007 opens after the carried entries, native to this issue: it bets on the metric nobody publishes, and on whether the good news repeats when somebody counts again.
Anthropic's ID-verification policy takes effect and, whatever its stated intent, functions in practice as a citizenship-sorted access path: US consumers regain restricted-tier access first, with no announced parity path for Indian or GCC passport holders. Anthropic says the change is an unrelated appeals update; we forecast the observable outcome and will score it.
Scored 9 July 2026. Fable 5 came back for every consumer on earth on the same day, 1 July, because the US Commerce Department lifted the export controls on 30 June. The restoration ran through diplomacy, not identity checks, and it landed a week before the ID policy took effect on 8 July. The policy itself verifies identity and age for flagged consumer accounts, carries no nationality component at all, and exempts Team, Enterprise and API customers. The disconfirming evidence we carried inside the entry, Anthropic's statement that this was an unrelated appeals update, held up better than our forecast did. To score this a hit we needed restricted access re-sorted by passport through the verification flow. It was not.
At least one further US frontier-model release goes through government pre-release review rather than open launch, extending the pattern already visible in June.
Scored 16 July 2026, a day early, because the pattern resolved ahead of the date. OpenAI previewed GPT-5.6 with the US government for about a month, released it on 26 June as a limited preview to around 20 government-approved organisations, and only opened it to the public on 9 July after a federal evaluation window under Executive Order 14409's voluntary pre-release framework. That is government pre-release review rather than open launch, exactly as forecast. The honest complication belongs on the record: the White House publicly denied giving any green light, approval or clearance, and EO 14409 explicitly bars mandatory licensing or preclearance. The claim required review, not approval; review demonstrably happened, so the hit stands on the wording as published.
At least one of TCS, Infosys, Wipro or HCLTech publicly announces a formal multi-model or sovereign-fallback architecture policy as strategy, not as a procurement footnote.
still trending toward a hit and still not scored early. No qualifying announcement from any of the four in the fortnight to 13 August; company newsrooms were checked directly. TCS's only release in the window was a single-vendor Google Cloud Gemini experience centre in Mexico, which points the other way. The nearest precursors remain HCLTech's digital sovereignty data-centre framing of 13 July and TCS Sovereignty Consulting from May. Scoring on 30 September.
At least three of India's top four IT firms disclose a named AI-revenue metric, in whatever form each chooses, in their Q2 FY27 results.
India's top four IT services firms, TCS, Infosys, Wipro and HCLTech, in aggregate add net headcount over FY27's first half, April to September 2026, while each scales AI-attributed revenue, confirming the reroute: the work returns offshore even as the Western rhetoric softens.
TCS alone added a net 9,279 in the June quarter on analyst arithmetic, reported as its largest sequential addition in fifteen quarters, with annualised AI revenue of $2.6bn. That is one firm and one quarter of the two-quarter window, and the entry requires the aggregate.
The first senior role explicitly titled for AI sovereignty or model continuity, distinct from CISO or Chief AI Officer, is publicly posted by a GCC entity or Gulf sovereign-linked employer.
nothing qualifying through 13 August, and a criteria risk is now worth stating in advance. Every relevant appointment we can find sits in the Chief AI Officer family, including Dubai's 22 government CAIOs. If the sovereignty mandate is being absorbed into CAIO roles rather than generating a distinct title, this entry will resolve as a definitional miss rather than a real-world one, and we would rather say so before the date than after it. The negative finding is also weak by construction, since Arabic-language decrees are under-indexed in the sources we can reach.
At least one multinational publicly names the Philippines, Romania or Poland, India's closest challengers on this index, as the lead location for a new AI-delivery or engineering hub, chosen over India, in a 2026 announcement.
On the next annual refresh of this index, India retains first place on the outsourcing-led composite while staying outside the top three on the capability-weighted view, confirming that its lead rests on delivery scale rather than AI preparedness.
MoHRE publicly adjusts, delays or waives an element of Emiratisation enforcement, citing market conditions, before 31 December 2026.
unchanged and still heading for a miss. Nothing from MoHRE in the fortnight to 13 August. The most recent substantive posture remains the July statement that 95% of mandated companies met their first-half targets, which is compliance-positive and cuts against the forecast, and the Dh10,000 monthly per-role fines have been live since 1 July. Four and a half months remain, so it is not scored, but we continue to expect a miss.
At least one company that attributed 2026 layoffs to AI is publicly reported to have rebuilt the same function in India, the Gulf or Africa, directly or through a capability centre or outsourcing partner, within twelve months of the cut.
A Gulf sovereign-linked or government entity publicly launches an initiative to position the UAE or Saudi Arabia as an AI-work delivery hub, not only a buyer or funder of AI, consistent with the capability-strong, labour-light profile the index assigns the Gulf.
The Karnataka Platform Based Gig Workers Act survives its constitutional challenge, meaning validity upheld, or the petitions dismissed or withdrawn, by 31 March 2027.
Uber's petition, WP 19776/2026, was tagged into the pending IAMAI batch on 28 July, so the challenge is consolidating rather than fragmenting, which we read as mildly supportive.
At least two further UAE government entities, emirate-level or federal and excluding Ajman, complete and publicly announce a fully autonomous end-to-end government transaction by 31 March 2027.
Basis: Ajman's live precedent, its 100-initiative three-year executive phase with coordinators now appointed across entities, and the federal directive to convert 50% of federal operations, procedures and services to agentic AI within two years with 80,000 employees in training. Against it: Ajman's own flow retains a customer approval step, fully autonomous is a description governments apply generously, and a first-of-its-kind claim is easier to make once than to repeat with the same language.
no second entity has announced. The 9 August strategic-track launch is movement in the machinery rather than in the outcome this entry tests.
A further institutional tally of AI-related hiring against AI-related job losses in India, from Nomura or any other bank, consultancy, industry body or official source, published by 31 March 2027, again reports hires exceeding losses.
Basis: the flow that produced the first result is still running, with TCS adding a net 9,279 in the June quarter, Cognizant's first Frontier cohort due by the fourth quarter, and 64% of new global capability centre roles created in 2026 requiring AI, data or automation skills, while the elimination side is concentrated in support functions already well through their automation. Against it, and this is a criteria risk we would rather state now than at resolution: the original is anecdote-count methodology and highly sensitive to which episodes a compiler happens to collect, one large Indian IT redundancy round would swing it, and no institution has committed to repeating the exercise at all. If no qualifying tally is published by the date, we score this a miss and say plainly that it failed for want of a publication rather than for want of the phenomenon.
Kenya enacts its AI policy, or an AI Bill, with the data-worker pay provision substantively intact, meaning pay for annotation, moderation or evaluation work calibrated against international rates for equivalent work, by 30 June 2027.
still nothing. No ministry statement, submission count, revised draft or industry response in the fortnight to 13 August, ten days after the consultation was set to close on 4 August. The Digital Policy Alert tracker still shows the measure as processing consultation with no event recorded after 4 August. Kenya's own visible activity this fortnight was on the growth side rather than the protection side, with a cooperation agreement signed on 4 August to expand global business services.
TechCabal Insights' full-year 2026 tracker records African tech layoffs above the half-year record of 2,574, while still naming AI as a direct cause in under 10% of tracked events: the cuts scale and the attribution does not.
Basis: the H1 record was driven by restructuring and banking consolidation that has not concluded, and the offshore losses that are genuinely AI-driven are decided by foreign clients who file nothing locally, so they cannot enter the tracker at all. Against it: a single large agent-deployment redundancy at a named African employer, of the Zap Africa kind but larger, would move the attribution share quickly off a small base.
None of India's top four IT services firms, TCS, Infosys, Wipro or HCLTech, publishes a redeployment rate by 30 June 2027: that is, any disclosed metric giving, for a defined period, the share of employees whose roles were automated or eliminated in favour of AI who remain employed by the firm, and in what function.
Basis: this is the metric that would settle whether a positive net headcount represents a transition or a replacement, and no firm in any market in this issue discloses it. The four publish quarterly headcount, attrition, AI revenue run rates and training and certification counts, none of which distinguish a redeployed worker from a new hire, and no regulator anywhere requires the figure. Nomura's own finding, that displaced workers rarely transition to AI engineering roles, is precisely what makes the disclosure unattractive. Against it: human-capital reporting in Indian IT is genuinely competitive, the figure would cost little to compute for a firm whose number is good, and a single firm choosing to differentiate on it would resolve this entry immediately.
Accenture's new bookings decline year on year again, confirming the June repricing as structural rather than sentiment.
Challenger's AI-attributed US job-cut count for the second half of 2026 exceeds the first half's 101,743, despite the softened executive rhetoric. The narrative and the number diverge further, not less.
the July figures are now verified against Challenger's own publication rather than wire coverage, closing the debt Issue 020 opened. July ran 10,970 AI-cited against 112,713 year to date, about 24% of all cuts, with AI leading all reasons for a fifth straight month. The H1 baseline of 101,743 is confirmed by the June report but is not stated in the July one, where it is exactly implied by subtraction; we cite it as derived when quoting the July document. One month of H2 at July's rate leaves the second half short, so the entry needs an acceleration it has not shown. Next data point is the August report at the start of September.
Data & AI remains a top-two growth sector in every remaining 2026 quarterly Cooper Fitch Gulf Employment Index, even if total GCC hiring stays flat or negative.
A monthly board-ready brief tying AI to the workforce and location decisions in front of you. Built by reAImagine.work and InGovern Research Services. Bengaluru and Dubai.
What you receive
A board-ready brief tied to your sector and the workforce decisions in front of you this quarter, this fiscal year, and twelve-plus months out.
Who it's built with
Editorial by reAImagine.work, founded by Debu Mishra. Board governance practice from InGovern Research Services, founded by Shriram Subramanian. Bengaluru and Dubai.
How it lands
Approve. Appoint. Commission. Separate. Deploy. Decide. Establish. Stakes in INR or USD ranges. No theatre.