AI and the future of work describes how artificial intelligence changes which tasks people perform, how companies organise teams and how software businesses earn money. Canva’s November 2026 restructuring is one of the clearest recent examples. The Australian design platform said 1,271 employees, about 22% of its workforce, would be affected through job losses, role changes and relocation.
Many people want to know whether AI caused these cuts. The evidence does not support that conclusion. Canva named slower growth, changing technology and organisational complexity as its reasons.
This article explains what the company changed, what the case shows about the impact of AI on jobs and what business leaders can apply to their own organisations.
What Did Canva Announce, and Did AI Cause the Job Cuts?
Canva’s announcement followed its first growth slowdown in a decade. CEO Melanie Perkins also pointed to changing technology and shifting customer expectations. The company did not attribute all 1,271 affected positions to AI, and no one should present it that way.
The more accurate reading is that AI formed one part of a wider business problem. A company that grew fast for ten years had accumulated layers, handovers and overlapping teams. AI exposed that weight because smaller teams can now produce more than they could before.
Why Are Software Companies Restructuring in the AI Era?
A software restructuring in the AI era is a deliberate redesign of teams, reporting lines and spending priorities in response to new technology and commercial pressure. Canva made three structural changes:
- Combining teams: Product, Growth and Platform functions will join one Product organisation.
- Reducing organisational layers: The company will consolidate narrow specialisations and overlapping responsibilities.
- Centralising collaboration: Sydney will become the main base for more product teams, and some employees will need to relocate.
Canva says these changes will reduce handovers, improve accountability and give smaller teams more ownership of their products.
Software firms are not simply adding automation to existing departments. They are asking whether their structures still suit a world where small teams can do more.
Atlassian offers a second example. Reuters reported that it announced 1,600 job cuts in March 2026 while shifting investment towards AI and enterprise sales. The two companies have different reasons, yet both are rebalancing their workforce and spending.
How is AI Changing the Traditional Software Business Model?
The traditional software business model sells specialised applications, usually priced per user. Companies bought separate tools for design, project management, customer relationship management, analytics and communication. Each tool solved one operational need.
Generative AI weakens that arrangement. A marketing professional can now use one AI assistant to write campaign copy, create visual concepts, analyse performance data and prepare a report in a single connected workflow. Traditional software will not disappear, but the source of its value is moving from access to features towards delivery of complete outcomes.
| Old model | Emerging model | What changes for the customer |
| Features | Workflows | Software completes connected tasks, not isolated functions |
| Subscriptions per seat | Outcomes | Buyers value measurable productivity gains more than seat counts |
| Standalone applications | Integrated systems | AI agents reduce the need to move information between tools |
One consequence deserves more attention than it usually receives. Seat-based pricing assumes that more users mean more revenue. If one person with an AI assistant does the work of three, the vendor sells fewer seats unless it changes how it charges. In my assessment, pricing is the part of this shift that software leaders discuss least and will struggle with most.
Does AI Make Software Companies More Profitable?
Not automatically. AI productivity does not translate directly into higher margins, because every AI feature carries running costs. These include computing infrastructure, model development, inference capacity, data management, security and ongoing maintenance.
The cost problem is sharpest for companies with large free user bases. Each AI request from a non-paying user costs money to serve and returns no subscription revenue. Forbes Australia reported that Canva delayed the rollout of some products because of the cost of serving AI features to its largely non-paying users. Canva has also introduced its own foundation model, Craft, to improve efficiency and open new revenue opportunities.
This is the core of AI-driven business transformation: leaders must weigh the customer value AI creates against the cost of delivering it. Adding an AI feature is a product decision. Building a sustainable AI business requires a cost structure that works at scale.
Will AI Replace Jobs or Reshape Existing Roles?
For most software roles, AI is more likely to reshape the work than remove it outright. Tasks inside a job can be automated while the job itself continues to exist with a different mix of skills.
Software development shows the pattern. AI coding assistants help developers write code, find errors, create tests and document applications. Businesses still need people who understand system architecture, security, performance, customer needs and the consequences of releasing software at scale.
| Function | What AI handles | What people still own |
| Design | Visual concepts and layouts | Creative direction, brand judgment, user understanding |
| Marketing | Campaign variations and data analysis | Positioning and audience priorities |
| Customer support | Routine queries | Complex complaints and sensitive decisions |
| Product management | Research and documentation | Deciding which problems deserve investment |
| Quality assurance | Automated defect detection | Judging reliability and real-world performance |
So. the useful question is not whether machines will replace people. The useful questions are how much of each job can be automated, which new responsibilities appear and how quickly employees can adapt.
Workforce planning also needs care here. A company that measures productivity only by tasks completed can overlook the judgment, coordination and accountability that make a product dependable. Those qualities rarely show up in a task count, which is why they tend to disappear first in a poorly planned cut.
Why Does Organisational Design Matter in AI-Driven Companies?
Organisational design is the way a company divides work, authority and decisions among its people. Canva’s restructuring shows that adopting new technology inside an outdated structure limits the benefit.
Growing companies create specialised departments to handle complexity. Specialisation builds expertise, but it also adds approvals, communication gaps and competing priorities. Canva’s decision to merge Product, Growth and Platform targets that problem directly. Its leadership named excessive handovers and fragmented ownership as obstacles to building a unified product.
For companies pursuing AI workforce transformation, efficiency depends on how work is distributed, not only on headcount. Smaller teams can offer:
- Faster decisions with fewer approval stages
- Clearer accountability for product outcomes
- Less duplication between departments
- Better coordination between technical and commercial functions
- Greater ownership from development to delivery
Fewer layers do not guarantee better results. Smaller teams still need enough expertise, clear decision rights and realistic workloads. Removing people without redesigning responsibilities raises pressure on those who remain and creates operational risk. Better design should be the goal, and a lower headcount is not proof of success.
A practical test for any leader: count how many separate teams must approve a typical product change. If the number is high and nobody can name a single owner of the outcome, the structure is likely the problem, regardless of what tools the company adopts.
What Can Business Leaders Learn from Canva’s Restructuring?
Canva’s case offers five lessons for executives dealing with AI and the future of work.
1. Keep revenue growth and organisational growth connected: A company can add staff, products and offices while becoming harder to run. Canva’s slowdown came after a decade of expansion, which suggests that structures built for one stage of growth may not suit the next. Leaders should check regularly whether complexity supports growth or slows execution.
2. Require a clear economic case for AI investment: Enterprise AI adoption should be judged against measurable business results. Leaders should account for:
- The cost of developing or licensing AI systems
- Infrastructure and computing expenses
- Training and integration costs
- Productivity gains that can be measured independently
- Revenue opportunities from new capabilities
An AI feature that attracts users but costs more to run than it earns weakens the business.
3. Treat simplification as more than cutting staff: Canva merged teams and consolidated overlapping responsibilities. Leaders should find repeated approvals, duplicated duties and unnecessary handovers first, and only then decide which roles need to change.
4. Recognise that access to AI is not an advantage by itself: As AI tools spread, adopting them sets a company apart less and less. Proprietary data, customer relationships, strong distribution, trusted products and deep market knowledge matter more. For the future of software companies, execution quality and the ability to solve specific customer problems will separate winners from the rest, because competitors can license similar models.
5. Weigh the long-term cost of workforce decisions: Layoffs reduce immediate costs but also remove institutional knowledge and specialist skill. Before restructuring, leaders should identify which capabilities their future products and customer relationships depend on.
Canva’s support package shows the human side of these decisions. The company announced at least six months of salary, extra healthcare coverage in eligible countries, career assistance and US$10,000 grants for departing employees who want to pursue independent projects. Leaders carry a responsibility to the people who built the organisation, and the quality of that support affects trust among those who stay.
Case Study: How Accenture Used AI to Improve Software Development Without Replacing Its Engineering Teams
Example: AI-assisted software development | 2024–2025
Accenture offers an interesting example of how companies can use AI to improve workforce productivity while retaining developers for complex engineering responsibilities.
In 2024, Accenture partnered with GitHub to evaluate how its developers could use GitHub Copilot in everyday software development. Rather than immediately deploying the tool across the organisation, the company began with a pilot involving 20 developers before expanding its adoption.
The subsequent enterprise research examined developers using Copilot alongside a comparison group without access to the tool. The study reported several improvements:
- Faster development: Developers could generate code, find solutions and complete repetitive programming tasks more efficiently.
- Better workflow: AI assistance reduced time spent searching for examples and handling routine coding activities.
- Improved developer experience: 95% of surveyed developers reported enjoying coding more with Copilot.
- Higher adoption: GitHub’s research reported that 67% of participating developers used Copilot at least five days per week.
- More successful builds: The enterprise study reported an 84% increase in successful builds among the participating developers.
What can businesses learn from this?
Accenture’s experience illustrates an important distinction: AI can increase the output of existing employees without requiring businesses to eliminate the roles themselves.
Developers could spend less time writing repetitive code and more time on architecture, problem-solving, testing and customer requirements. The organisation also invested in training, onboarding and internal knowledge-sharing to support adoption.
The results should be interpreted carefully: these are findings from a GitHub-led study, not proof that every company will achieve the same productivity improvements or that AI adoption has no employment consequences.
The leadership lesson is that before deciding which positions to eliminate, companies should examine whether AI can help existing employees perform higher-value work. Accenture’s example demonstrates why workforce transformation can begin with redesigning tasks and workflows rather than simply reducing headcount.
How Can Companies Prepare Their Workforce for AI?
Preparing a workforce for AI means linking technology spending to changes in what people actually do. Buying software and running occasional training sessions does not achieve that.
A practical approach has 5 steps:
- Assess tasks, not jobs: Identify repetitive activities AI can automate and responsibilities that need human judgment.
- Reskill current employees: Teach AI tools, data literacy, automation and critical thinking.
- Redesign job descriptions: Combine relevant technical and business skills instead of creating narrow specialist roles.
- Set responsible AI standards: Define rules for data protection, accuracy, accountability and human oversight.
- Measure real productivity: Check whether AI improves output quality, customer experience and operating costs, and do not treat adoption rates as the result.
The impact of AI on jobs also depends on what a company does with the gains. It can use higher efficiency to expand, build new products or reduce workloads. Turning those gains into job cuts is a management choice and not an automatic result of automation. For executives, AI and the future of work calls for long-term planning across workforce development, employee skills and business model change.
Conclusion
Canva’s restructuring shows how changing technology, slower growth and organisational complexity combine to reshape the software industry. It is not evidence that AI alone caused 1,271 employees to be affected. It does show the pressure on established companies to rethink how they operate. AI and the future of work will depend on more than automation. Sustainable growth requires better products, capable teams, sound economics and leaders who balance efficiency with human expertise.
Author’s Note:
In my opinion, Canva’s restructuring highlights five important lessons about AI and the future of work:
1. AI should improve productivity, not become an excuse for workforce reduction. Companies need to examine whether technology is creating new opportunities or simply being used to justify smaller teams.
2. Organisational problems cannot always be solved through automation. Canva’s decision to consolidate teams highlights how fragmented responsibilities, unnecessary handovers and management layers can slow down businesses even before AI enters the picture.
3. Productivity does not automatically translate into profitability. AI infrastructure, computing resources, implementation and maintenance involve costs. Businesses must understand the economics behind AI adoption before expecting financial gains.
4. Human expertise will remain a competitive advantage. Creativity, critical thinking, customer understanding, decision-making and accountability are capabilities that companies must continue to develop alongside AI-powered tools.
5. Leaders must rethink how productivity gains are distributed. If smaller teams can achieve more with AI, businesses should also consider investing in employee development, new products and better working practices rather than measuring success through headcount reduction alone.
In the end, I believe the companies that succeed will not necessarily be those that replace the most employees with AI, but those that understand where technology adds value and where human expertise makes the difference.
