# AI Enablement Work in 2026: Evidence from 100 Current Vacancies

> A bounded study of 100 current public vacancies across seven source families maps workforce AI literacy, learning, adoption support, responsible practice and evidence.

- Canonical page: https://mtfinstitute.com/insights/ai-enablement-work-100-vacancies-2026/
- Content type: Article
- Editorial category: Research & Reports
- Publisher: MTF Institute of Management, Technology and Finance
- Author: MTF Institute Research Team- Published: 2026-08-25
- Updated: 2026-08-25
- Language: English
- Topics: Responsible AI, Labour Market Research, AI Enablement, Workforce AI Adoption, AI Literacy, Champion Communities

## AI Enablement Work in 2026: Evidence from 100 Current Vacancies

The complete archive - a searchable PDF and the accepted-vacancy dataset - is preserved at [Zenodo DOI 10.5281/zenodo.22099514](https://doi.org/10.5281/zenodo.22099514).

**MTF Institute Research Team**  
Research cut-off: 25 August 2026  
Status: original research draft; not yet published

## Abstract

As organizations introduce generative and other forms of artificial intelligence into everyday work, a distinct workforce function is taking shape between AI strategy, technology delivery, governance, learning and organizational change. This report examines that function through a purposive corpus of 100 current, unique, first-party employer vacancies retrieved on 25 August 2026. Every accepted vacancy had to show at least three defined workforce-enablement responsibilities, including an explicit anchor in role-based AI literacy, learning and adoption programmes, or a champion community. Roles serving only product sales, customer success, engineering, general transformation or high-level strategy were excluded.

The findings show a coherent body of work beneath a fragmented set of titles. The corpus contains 82 unique exact titles. Only seven titles appear more than once, and the three most repeated labels—AI Enablement Lead, AI Adoption Lead and AI Enablement Manager—account for 17 records together. The role is therefore more standardized by responsibilities than by title. Employers repeatedly ask for learning programmes connected to real tasks, workflow discovery, role-sensitive literacy, responsible-use guidance, cross-functional handoffs, adoption evidence and continuous improvement. Champion networks appear as one important scaling mechanism, but not the only one.

The corpus does not establish total hiring volume, market growth, salary, productivity or business return. It does support a bounded professional interpretation: AI enablement is the design and operation of a workforce learning, practice, support and feedback system for AI uses that have already passed the relevant organizational decision process. This work does not replace AI governance, product management, legal review, human-resources authority, technical implementation or enterprise transformation strategy.

## Key findings

1. **The work is recognizable even though titles vary.** The 100 vacancies use 82 exact titles; 75 titles appear once. Forty-one titles contain *enablement*, 20 contain *adoption*, and 27 contain *transformation*. These overlapping markers show an emerging role family, not one settled occupational label.
2. **Learning is tied to work rather than awareness alone.** Eighty-nine records include learning and adoption programmes, 83 include task and workflow discovery, and 68 include role-based AI literacy.
3. **Enablement is a cross-functional operating responsibility.** Seventy-four records include handoffs across business, learning, technology, data, security, legal, risk or governance functions.
4. **Measurement is expected, but interpretation must remain cautious.** Seventy-one records include adoption, usage, confidence or value measures. Vacancy language describes employer expectations; it does not prove that any intervention caused a result.
5. **Responsible practice is part of rollout.** Sixty-five records include approved-use or responsible-practice duties. The enablement role translates decisions into teachable practice; it does not approve AI systems or determine legal compliance.
6. **The role combines process, learning and influence skills.** The most frequently coded skills are process mapping, communication and facilitation, learning design, stakeholder management, data and metrics, and change management.

## 1. Research question

The study asks: **What responsibilities and skills define workforce AI enablement in current employer vacancies, and where are the boundaries of that work?**

The question matters because the phrase *AI enablement* is used for several different activities. A software company may use it for customer adoption. A technology team may use it for infrastructure or developer support. A sales organization may use it for go-to-market training. A transformation office may use it as a broad label for an AI portfolio. Those uses do not automatically describe a workforce learning and adoption role.

This research therefore starts with duties, not titles. A record qualifies only when the public posting describes a material internal-workforce function or an explicit client-workforce adoption programme. The accepted work must connect learning, practice or a champion community to at least two other defined responsibilities. This rule prevents a title match from substituting for evidence about the job.

## 2. Method

### 2.1 Sampling frame

Researchers used purposive public-web queries for combinations of AI, enablement, adoption, literacy, learning, change, champions and workforce transformation. Candidate records had to resolve to a first-party employer page or an employer-controlled applicant-tracking page available without authentication on 25 August 2026. No access control, CAPTCHA, paywall, rate limit or technical restriction was bypassed.

The discovery process considered 276 candidate vacancies. After relevance, access and duplication checks, 100 were accepted and 176 were excluded. The final records came from seven source families:

| Source family | Accepted vacancies |
|---|---:|
| Workday | 59 |
| Greenhouse | 19 |
| Ashby | 12 |
| Lever | 7 |
| Workable | 1 |
| Employer careers site | 1 |
| SAP SuccessFactors | 1 |
| **Total** | **100** |

The source-family count describes the technical publication surfaces, not industries or geographic markets. Thirty-two accepted records did not expose a usable location in the indexed public posting. The report therefore does not construct a geographic distribution.

### 2.2 Inclusion rule

Each accepted vacancy had to satisfy all three conditions:

1. at least three distinct responsibilities from the ten-code taxonomy;
2. at least one anchor responsibility: role-based AI literacy, learning and adoption programmes, or a champion community; and
3. an internal workforce beneficiary or a clearly stated client-workforce adoption programme.

The ten responsibility codes were:

- task and workflow discovery;
- role-based AI literacy;
- approved use and responsible practice;
- champion community;
- manager and leader enablement;
- learning and adoption programmes;
- communication and change support;
- adoption, usage, confidence and value measures;
- cross-functional handoffs; and
- roadmap and continuous improvement.

Coding was conservative. A responsibility or skill was recorded only when the public text supplied an explicit evidence anchor. Short excerpts were retained as pointers, with a 20-word maximum, rather than as substitutes for the vacancy page. Skills were coded separately from responsibilities.

### 2.3 Deduplication and quality control

Canonical URLs were normalized. Deduplication used employer, title, location and canonical source URL in lower case. The accepted corpus contains 100 unique URLs, 100 unique composite keys and 97 employer names. Modaxo, PwC and TD Bank each contribute two distinct records; all other employer labels appear once.

The strict review tested required fields, URL and composite-key duplication, taxonomy validity, excerpt length, live public resolution and the inclusion rule. Eighteen initially flagged records received a second evidence review. Five were repaired only where additional public text explicitly supported the missing coding; 13 were removed and replaced. A further thin coding case was repaired from explicit role-based learning text. The final corpus has no strict-rule violation. Each record contains between three and nine coded responsibilities, with an average of 6.52 and a median of seven.

### 2.4 Interpretation rule

Because the corpus is purposive, every percentage in this report is a description of these 100 accepted records. It is not an estimate of all vacancies, employers or workers. Counts should be read as evidence of recurring operating patterns inside a tightly defined sample, not as market prevalence.

## 3. An emerging title family

The exact job title is unusually unsettled. The corpus contains 82 unique exact titles. Seventy-five appear once, while only seven appear more than once. The most frequent exact labels are:

| Exact title | Records |
|---|---:|
| AI Enablement Lead | 7 |
| AI Adoption Lead | 5 |
| AI Enablement Manager | 5 |
| AI Adoption & Enablement Lead | 2 |
| AI Transformation Lead | 2 |
| AI Transformation Manager | 2 |
| Director, AI Transformation | 2 |

Title markers overlap. Forty-one titles contain *enablement*, 20 contain *adoption*, 27 contain *transformation*, eight contain *change*, seven contain *learning*, three contain *training*, three contain *literacy*, and two contain *workforce*. Seniority language is also mixed: 36 titles contain *lead*, 25 contain *manager*, 18 contain *director*, three contain *consultant*, and one contains *head*. These markers are simple, case-insensitive text observations and are not mutually exclusive.

The pattern has two implications. First, the exact title *AI Enablement Manager* is defensible as a descriptive occupational label because multiple unrelated employers use it. Current examples include Crowe [AEM-006], Dodge & Cox [AEM-007], O.C. Tanner [AEM-036] and Systemiq [AEM-076]. Second, the label cannot define the research population by itself. Closely matching work appears under AI Adoption Lead at Marks & Spencer [AEM-079] and Sonova [AEM-081], AI Learning & Curriculum Lead at BMO [AEM-084], and Director of AI Upskilling and Reskilling at Mastercard [AEM-033].

The role family is thus better described by its work product: a system that helps defined workforce groups learn, practise, use, review and improve already-authorized AI-supported work.

## 4. Responsibilities observed in the corpus

The responsibility counts below are non-exclusive. A vacancy may contain several codes.

| Responsibility | Records | Share of this corpus |
|---|---:|---:|
| Roadmap and continuous improvement | 96 | 96% |
| Learning and adoption programmes | 89 | 89% |
| Task and workflow discovery | 83 | 83% |
| Cross-functional handoffs | 74 | 74% |
| Adoption, usage, confidence and value measures | 71 | 71% |
| Role-based AI literacy | 68 | 68% |
| Approved use and responsible practice | 65 | 65% |
| Communication and change support | 63 | 63% |
| Champion community | 41 | 41% |
| Manager and leader enablement | 2 | 2% |

### 4.1 Learning connected to tasks and workflows

The combination of learning programmes in 89 records and task or workflow discovery in 83 is central. Employers are not merely asking someone to explain what AI is. They are asking the role to identify where approved tools meet actual work, then design learning, practice and support around those contexts.

Marks & Spencer combines role-specific coaching, workflow use cases, local champions, responsible-use reinforcement and value tracking [AEM-079]. HP connects work mapping, output testing, user training, impact reporting and handoffs to legal, security and compliance partners [AEM-086]. Tekion describes ownership of a company-wide AI enablement programme linked to workflow discovery, literacy, champions and evidence [AEM-001]. These examples differ by industry and title but share a task-to-learning connection.

For practice, this means a generic AI awareness session is an incomplete unit of work. The enablement function needs to know which role is acting, which task is in scope, which tool or use has already been authorized, what human judgment remains necessary, what evidence must be checked, and where uncertainty or suspected misuse goes next.

### 4.2 Role-based literacy

Role-based AI literacy appears in 68 records. The coding requires explicit differentiation by role, persona, learner group or work context; general AI communication is not enough. Mastercard asks for learning experiences differentiated by persona and role [AEM-033]. Cencora connects enterprise AI fluency across roles with scalable learning and communities of practice [AEM-032]. Norton Rose Fulbright differentiates training by roles and user groups [AEM-031].

This finding supports a layered design. A common foundation may cover basic concepts, source awareness, confidentiality, uncertainty and human review. Role-level learning then adds the decisions, data conditions, common errors and escalation routes of a particular work context. Manager support adds coaching and work-system observation without turning learning records into employee ranking.

The evidence does not support one universal literacy level. Roles face different tools, information, consequences and review duties. A useful literacy map therefore describes observable work outcomes and prerequisites, not a single score presented as proof that every learner is competent.

### 4.3 Responsible use as part of adoption

Approved-use and responsible-practice duties appear in 65 records. This does not make enablement a governance or legal function. It shows that employers expect learning and rollout to carry decisions made elsewhere into everyday practice.

At Systemiq, role-spanning training, champions, use-case prioritization and responsible adoption appear together [AEM-076]. At Sonova, responsible-use education is connected to role-based learning, manager support, workflow adoption and measurement [AEM-081]. At BMO, responsible AI learning is embedded in use-case curricula and programme outcomes [AEM-084].

The practical boundary is important. An enablement professional can translate an approved policy into teachable examples, prohibited-input warnings, review steps, stop conditions and escalation routes. That professional should not classify the legal risk of a real AI system, authorize a tool, approve an exception or issue a compliance opinion. Those decisions remain with the appropriate owners.

### 4.4 Handoffs and organizational coordination

Cross-functional handoffs appear in 74 records. AI-supported work crosses learning, operations, technology, data, security, privacy, legal, risk, governance, communications and business leadership. The enablement function therefore operates as a connector, not an isolated training producer.

Crowe's public role description places the AI Enablement Manager between internal users, enablement resources and technology partners [AEM-006]. O.C. Tanner connects the function to leaders, adoption, responsible practice, measurement and technical partners [AEM-036]. Sonova connects the human capability layer to multiple control and business functions [AEM-081].

A practical handoff should record the question, the evidence available, the accountable receiving function, the expected response and the effect on learning content. If a policy changes, a tool feature changes, a learner reports unexpected output, or a use case becomes unclear, the learning programme needs a controlled route to update or pause its examples.

### 4.5 Champions, communication and manager support

Champion communities appear in 41 records and communication or change support in 63. Champion networks can extend peer support, surface recurring questions and provide local context. They should not become informal approval bodies. Dodge & Cox links a champions network to manager programming and broader organizational reach [AEM-007]. GHJ connects role-specific training, internal champions, communication and adoption indicators [AEM-060].

Only two records carry the narrow `manager_and_leader_enablement` code. This should not be read as evidence that managers matter in only 2% of roles. It reflects a strict coding threshold for explicit manager-focused enablement. Other records may mention leadership sponsorship, stakeholders or communication without meeting that narrower code. The finding is a reminder to distinguish the presence of managers as sponsors from a designed manager-support service.

### 4.6 Measures and continuous improvement

Seventy-one records include adoption, usage, confidence or value measures, and 96 include roadmap or continuous improvement. Employers expect the function to observe whether learning reaches the intended groups, whether supported practices occur, where confidence or friction changes, and which content needs revision.

These measures have different meanings. Completion shows access to a learning activity, not competent workplace use. Usage shows activity, not quality or benefit. Confidence is a self-report, not proof of correctness. A quality observation may reveal a process issue but does not by itself identify its cause. Business-value language in a vacancy is an expectation attached to the role; it is not evidence that a particular course or tool produced a return.

A defensible evidence chain links a stated learning outcome to an opportunity to practise, an aggregate observation, an interpretation limit and a review action. Small groups should be suppressed or combined when reporting could identify individuals. Measures should improve the learning and support system, not create employee surveillance or performance scores.

## 5. Skills employers connect to the work

The corpus codes ten skill families separately from responsibilities:

| Skill | Records |
|---|---:|
| Process mapping | 73 |
| Communication and facilitation | 71 |
| Learning design | 67 |
| Stakeholder management | 67 |
| Data and metrics | 64 |
| Change management | 60 |
| Governance and risk fluency | 56 |
| Programme management | 53 |
| AI fluency | 49 |
| Organizational design | 18 |

The combination is more informative than any single count. Process mapping helps connect learning to real tasks. Learning design converts those tasks into progressive outcomes, practice and feedback. Communication and stakeholder skills help the function work across roles and translate specialist decisions. Data skills support cautious measurement. Governance and risk fluency helps the professional recognize boundaries and route questions without taking over specialist authority. AI fluency supports credible examples and evaluation of changing tools, but the role is not defined as software engineering.

Skills were coded only when the public text supplied an explicit anchor. A low count therefore means “not explicitly coded in this corpus,” not “unimportant.” The corpus also does not establish a universal entry route. Employers may recruit from learning and development, organizational change, programme delivery, consulting, digital transformation, operations or technology. The repeated requirement is the ability to connect workforce practice with authorized AI use and observable learning support.

## 6. A bounded operating model for workforce AI enablement

The evidence supports seven practical work products. These are original descriptive tools, not reproductions of an external method.

1. **Role–Task–Use–Review Inventory.** Record the role context, task, already-authorized AI use, required human judgment, evidence and escalation route. This is not a system inventory or automation-opportunity register.
2. **Role-Based Literacy Coverage Map.** Connect role contexts to observable outcomes, prerequisite knowledge and practice evidence. This is not a vendor syllabus, external skills framework or legal standard.
3. **Policy-Translated Use Pattern Card.** Turn an attributed policy or governance decision into a teachable scenario with permitted inputs, prohibited inputs, review, stop and escalation fields. The card carries approval status; it does not create it.
4. **Human–AI Review Rehearsal.** Let learners practise source checking, uncertainty, omission detection and escalation in a fictional case. It is not a branded prompt formula.
5. **Champion Support Loop.** Define peer-support roles, office hours, question capture, escalation and content feedback. Champions support learning; they do not approve uses or exceptions.
6. **Enablement Evidence Chain.** Link an outcome, practice opportunity, aggregate observation, interpretation limit and improvement action. It is not assurance or causal proof.
7. **90-Day Workforce Enablement Roadmap.** Sequence role cohorts, learning releases, dependencies, accessibility, support and review. It is not an enterprise transformation plan or technical deployment plan.

Together, these products form a Workforce AI Enablement Plan. The plan begins after the organization has supplied the relevant strategy, tool decisions, policies and specialist constraints. It ends with learning, practice, support, evidence and feedback routed back to the responsible owners.

## 7. Portfolio and role boundaries

The strict inclusion process shows why boundaries matter. Thirteen initially accepted records were removed when closer review found customer adoption, generic transformation, technical delivery or insufficient learning evidence. Examples included a vendor adoption strategist, a generic change director, several transformation managers and a customer-oriented enterprise adoption manager. Replacements had to satisfy the unchanged three-duty, anchor and beneficiary rule.

The resulting role is distinct from adjacent professional territories:

- **Individual manager productivity:** using prompts and templates for one's own meetings, reports, objectives or career material is not a workforce enablement system.
- **AI governance:** inventories, risk routing, control design, formal approval, incidents and assurance belong to governance and specialist owners. Enablement consumes their decisions as inputs.
- **AI product management:** discovery, feasibility, requirements, evaluation claims and release decisions precede enablement for an approved product or tool.
- **Enterprise transformation:** investment choices, platform strategy, operating-structure redesign and portfolio benefits sit above the enablement workstream.
- **Human resources and employment decisions:** hiring, promotion, termination, compensation, workforce reduction, performance rating and employee relations remain outside scope.
- **Automation and engineering:** workflow redesign, bots, integrations, agents, deployment and technical optimization are not learning-programme responsibilities.
- **Sales, partner and customer success:** these roles qualify only when the posting clearly describes a client-workforce learning and adoption programme, not account growth or product utilization alone.

These boundaries protect both the integrity of the role and the people affected by it. They also prevent a course or certificate from implying authority that the research does not support.

## 8. Legal, rights and claims boundaries

Public vacancies are factual demand evidence, but their wording remains protected. This study retains employer, title, date, URL, coded facts and short evidence pointers. It does not reproduce vacancy bodies or imply employer endorsement.

Official legal and government materials can inform context, but a workforce plan cannot prove legal compliance. Requirements vary by jurisdiction, system, role and date. Any real implementation needs current review by the appropriate legal, privacy, security, human-resources, accessibility and worker-representation functions. Learning content should route unresolved questions rather than decide them.

Employee-level analytics require particular caution. The enablement function should not design candidate ranking, promotion or termination recommendations, individual productivity surveillance, emotion or health inference, automated performance reviews, or individual AI-readiness scores used for employment consequences. Learning evidence should be proportionate, access-controlled and aggregate by default, with fictional data used in instructional cases.

The work also does not require proprietary change, evaluation, skills or certification systems. Course and workplace tools can be authored from first principles using ordinary descriptive language. Content from ISO publications, Prosci, Kotter, Kirkpatrick, SFIA, vendor academies, certification objectives or competitor courses should not be copied or used to imply external recognition without a specific, documented licence.

Finally, the corpus supports no promise of employment, promotion, salary, adoption, productivity, cost reduction, business return, performance, legal status or external recognition. It identifies responsibilities that employers describe in current vacancies. It does not prove what a learner, employer or technology will achieve.

## 9. Practical implications

### For employers

Define the function by outcomes and handoffs before choosing a title. A useful role brief should state which workforce groups are served, which AI uses have already been authorized, how learning connects to tasks, where responsible-use questions go, how champions are bounded, what aggregate evidence will be collected, and who owns updates. Separating enablement from governance, product and employment authority reduces ambiguity.

### For learning and development teams

Move beyond one universal awareness session. Build a common foundation, then add role-sensitive outcomes, sanitized practice, manager support where explicit, peer support and controlled content updates. Treat completion, usage, confidence, quality and value as different measures with different limitations.

### For AI, data and technology teams

Provide enablement with stable inputs: approved tools and uses, relevant release information, data-handling constraints, known failure patterns, review requirements and escalation contacts. When a tool or policy changes, the learning system needs a clear update path. Technical teams should not assume that access equals adoption or that a general demonstration prepares every role.

### For governance, legal, security and privacy teams

Express decisions in forms that can be translated into practice. A learner needs to know what inputs are allowed, what review is required, when to stop and whom to contact. The enablement function can make those instructions understandable and rehearse their use, while leaving formal authority with the responsible function.

### For professionals entering the role

Build evidence of integration rather than claiming broad AI leadership. A credible portfolio might include a role–task inventory, literacy coverage map, policy-translated scenario, learning route, champion support design, measure dictionary and 90-day roadmap. Each item should show source constraints, human review, confidentiality, interpretation limits and handoffs.

## 10. Limitations

This study is a bounded, purposive sample, not a census. Search terms favored roles that use recognizable AI enablement, adoption, literacy, learning and transformation language. Roles with similar work but different terminology may be absent. The corpus also reflects what employers chose to publish and what public search and employer pages exposed on one date.

Vacancies can close or change after retrieval. Thirty-two records did not expose a usable location, so the study does not compare regions. Applicant-tracking source families are reported, but they do not represent industries. Employer names are not weighted by workforce size, hiring volume or economic importance.

Responsibility and skill codes are analytical classifications. Conservative coding reduces unsupported inference but may undercount duties expressed indirectly. Counts are non-exclusive and do not show how much time a role spends on each duty. The two records coded for explicit manager and leader enablement illustrate this limitation: manager support may appear more broadly without meeting the narrow evidence threshold.

The study does not analyze salaries, filled positions, applicant counts, course sales, learning completion, productivity, business return or employment outcomes. It does not test whether the described practices work. It does not provide legal advice, rights clearance for every possible source, or an exhaustive occupational-title or trademark search.

## Conclusion

Across 100 current first-party vacancies, workforce AI enablement appears as a coherent operating function hidden beneath varied titles. Its core is not generic enthusiasm for AI and not technical implementation. It is the structured connection between authorized uses, real tasks, role-sensitive learning, responsible practice, peer and manager support, cross-functional handoffs, evidence and continuous improvement.

The fragmented title landscape makes duty-based definition essential. *AI Enablement Manager* is a defensible descriptive title, but it needs a clarifying workforce subtitle and explicit exclusions. The role should be understood as downstream of strategy, product and governance decisions and outside employment-decision authority.

For professional learning, the strongest outcome is therefore practical and bounded: a Workforce AI Enablement Plan that shows who needs to learn what, for which authorized work, with which practice and support, under which constraints, measured in which limited way, and connected to which accountable owners. That outcome reflects the work visible in the corpus without claiming that a course, title or tool guarantees adoption or business results.

## Reproducibility and selected public sources

The accompanying research archive contains the 100-record source ledger, machine-readable corpus, methodology and deterministic QA. Each accepted record preserves employer, title, location where available, public URL, retrieval date, source family, live-status evidence, a short evidence pointer, responsibility codes, skill codes and a composite deduplication key.

Selected first-party examples cited in this report:

- [AEM-001 — Tekion, Director, AI Transformation & Enablement](https://jobs.ashbyhq.com/Tekion/d624639e-bff3-4af0-ab3c-8ea810ce5707)
- [AEM-006 — Crowe, AI Enablement Manager](https://crowe.wd12.myworkdayjobs.com/en-US/External_Careers/job/AI-Enablement-Manager_R-51495)
- [AEM-007 — Dodge & Cox, AI Enablement Manager](https://dodgeandcox.wd5.myworkdayjobs.com/en-US/Dodgecox/job/AI-Enablement-Manager_R0000628)
- [AEM-031 — Norton Rose Fulbright, AI Transformation Change & Adoption Manager](https://nrf.wd3.myworkdayjobs.com/External/job/London-United-Kingdom/AI-Transformation-Change---Adoption-Manager_R-3740)
- [AEM-032 — Cencora, Director, Emerging Technology & AI Fluency](https://myhrabc.wd5.myworkdayjobs.com/en-US/Global/job/Director---Emerging-Technology---AI-Fluency_R268440)
- [AEM-033 — Mastercard, Director, Learning & Development — AI Upskilling & Reskilling](https://mastercard.wd1.myworkdayjobs.com/en-US/CorporateCareers/job/Director--Learning---Development---AI-Upskilling---Reskilling_R-273638)
- [AEM-036 — O.C. Tanner, AI Enablement Manager](https://octanner.wd501.myworkdayjobs.com/en-US/O_C_Tanner/job/AI-Enablement-Manager_JR26-131)
- [AEM-060 — GHJ, AI Change Management Lead](https://jobs.lever.co/ghj/35ae479f-d57a-4be3-818a-c7530c6e5627)
- [AEM-076 — Systemiq, AI Enablement Manager](https://job-boards.eu.greenhouse.io/systemiq/jobs/4747154101)
- [AEM-079 — Marks & Spencer, AI Adoption Lead](https://jobs.marksandspencer.com/job-search/digital-tech/london-greater-london/adoption-lead/300008056234887)
- [AEM-081 — Sonova, AI Adoption Lead](https://jobs.sonova.com/job/Berlin-AI-Adoption-Lead/1423492533/jobs.sonova.com)
- [AEM-084 — BMO, AI Learning & Curriculum Lead](https://bmo.wd3.myworkdayjobs.com/en-US/External/job/AI-Learning---Curriculum-Lead--Commercial-Banking-_R260011154)
- [AEM-086 — HP, AI Solutions and Adoption Lead](https://hp.wd5.myworkdayjobs.com/en-US/ExternalCareerSite/job/AI-Solutions-and-Adoption-Lead_3164411-1)

All public sources were retrieved on 25 August 2026. URLs are preserved for verification; later closure does not change the dated observation.


## Citation

When citing or summarizing this material, link to the canonical HTML page: https://mtfinstitute.com/insights/ai-enablement-work-100-vacancies-2026/