Graduates worry AI will take away advertising jobs. Are they right to be concerned?

In interviews and public discussions of the impacts of that investment, most have been quick to downplay the risk of job losses related to AI adoption. But a Forrester study released in June predicted that up to a third of US agency jobs could be lost by 2030.

Concerns about creative jobs being swallowed up by AI have already taken hold among graduates and school leavers considering advertising as a career path. A recent study by Ravensbourne University London found that 37% of 18- to 25-year-olds believed there wouldn’t be many creative jobs available in marketing because of AI.

Taken together, it’s clear there is a risk – and that it might be putting off people who might otherwise be heading towards a career in advertising. But it’s still unclear, as yet, which types of jobs are actually at risk and which might genuinely benefit or change but remain fundamentally intact.

We’ve taken the time to review some of the research published on this topic since the launch of ChatGPT in November last year, with the aim of answering those questions.

• Adland has a talent problem. Emphasis on tech and analytical roles can solve it

First of all, according to the Organization for Economic Co-operation and Development (OECD), the jobs most at risk from automated competition are highly skilled roles, particularly those involving financial, legal and cultural expertise.

Partially, that’s because those positions require a high degree of training (a law or accountancy degree, say) and command a high salary, the OECD’s Stijn Broecke writes. Though it was assumed that complex, knowledge-based roles would be safe from automation, recent developments have “exposed” those specialisms to AI advancements, which have “extended the potential scope of automation considerably beyond what had previously been possible.“

So, there are now folks in previously secure, well-paid jobs worried that their employers could begin to see them as expensive albatrosses around a company’s bottom line. In fact, Broecke notes, 20% of finance workers in seven OECD countries report that they’re “extremely worried” about job losses over the next decade.

‘Exposure’ to automation is something we’ll likely be hearing more about in the future. A University of Pennsylvania working paper released in March of this year measured exposure as “whether access to a large language model (LLM) would reduce the time required to complete a specific detailed work activity… by at least 50%.”

In plain English, if half of the tasks involved in a job could be done quicker with AI assistance, that job is considered to be exposed. Though lots of studies have used a similar measurement framework, there’s disagreement about the levels of exposure currently present within the economies of the largest advertising markets. That report’s author, academic Tyna Eloundou, estimates that “about 80% of the US workforce could have at least 10% exposure,” while a Goldman Sachs research paper published this year suggested that two-thirds of US and European jobs were exposed.

Exposure doesn’t have to mean a position will be removed, just altered. Studies from 2019 found that some exposure to AI was associated with wage growth because it increased the productivity of a given role and made it more valuable to employers.

Though it’s unclear whether or not it’s yet triggered wage growth in the ad sector, we know plenty of marketers have begun to boost their productivity through AI. A Hubspot survey released this year, for example, found that 45% of marketers were using gen AI tools to create content while 53% were using it to make edits to copy they’d written. 83%, its survey found, said AI helped them create significantly more content (email newsletters, social posts and corporate blogs) than they could without it. These developments have arrived without triggering major job losses.

Ultimately, both outcomes are possible. Forrester’s Agency AI-Powered Workforce Forecast, 2030 anticipates: “Some agency roles will be positively impacted by AI and automation while others will be negatively impacted, resulting in an overall workforce reduction.”

Finance and legal positions require a vast amount of knowledge, but they don’t necessarily involve personal judgment (creative bookkeeping isn’t usually a good thing). IBM, for example, paused hiring in May for human resources and back-office jobs that it expects could be replaced with AI.

But advertising gigs that do require creativity and judgment – even if they’re exposed to AI automation – could benefit. Forrester’s report predicts that “creative curator skillsets” will become key; putting editorial nous, personal taste, an eye for imagery and the unusual at a premium.

And technical roles, such as VFX animators and UX designers, could also benefit. Ravensbourne’s study, released in August as British A-level students began to confirm their higher education plans, found that young people weren’t aware of those roles, despite them being among the highest-paying jobs for the university’s most recent graduates.

As companies and customers begin to adopt AI tools as a means of engaging with and connecting to content and commercial spaces on the web, the stock UX expertise is going to rise. In fact, Forrester’s report anticipates the number of software developers and similar roles to increase as a proportion of the overall advertising workforce.

While it suggests finance (-0.2%), administrative (-0.5%) and clerk roles (-2.1%) will fall, jobs on the creative side – and the managers required to marshal them – are predicted to increase by 0.6% and 1.4% respectively.