Read time: 5 minutes

McKinsey recently released their annual report on the State of AI based on a survey of over 1,600 global execs, proving a snapshot of look at how companies are leveraging both established and emerging AI technologies. Here we look at the key point and suggest routes for businesses to approach AI.

GenAI Adoption Reaches an Inflection Point

According to the report, a third of organizations are using generative AI tools regularly in at least one business function. The most common uses are in marketing and sales, product development, and customer service - and 40% of companies that are already using AI will increase investments because of generative AI.

Nearly a quarter of C-suite execs said they are personally using gen AI tools for work, however, just 21% have formal policies governing gen AI use and only less than half are mitigating risks like inaccuracy.

AI-Driven Companies Widen Their Lead

The research identified a segment of high-performing companies where AI contributes over 20% of EBIT. These organizations adopt generative AI far more aggressively than others, especially for revenue-generating use cases. They invest around 5x more in AI overall than their peers. 

Even with these more aggressive companies only a third are leveraging best practices like reusing AI model components versus reinventing them. 

The Need for Reskilling Outweighs Job Losses

A very common fear of AI is that it will displace jobs (for workers), while respondents predict more employees will be reskilled than lose their jobs over the next 3 years. 

Certain sectors will bear more of the brunt of any losses - technology and financial sectors are mentioned. In some cases over 20% of staff will need reskilling due to AI adoption. There is also the impact of which type of jobs and roles will be impacted - while white collar roles will be in the mix, there is a also a potential gender split of impact to consider.

Companies will need to be rethinking talent strategies, workflows, and culture to make any augmentation successful - but also to keep existing employees and workers on-side.

While hiring general tech talent has become perhaps a little easier with current tech sector cutbacks, most report ongoing challenges in acquiring specific AI talent (perhaps an output of a lack of STEM focussed compulsory training in schools?). Creative approaches like upskilling programs, gig work, crowdsourcing, and AI tools themselves will be required to access the expertise needed to scale AI.

Overall AI Adoption Remains Steady But Narrow

55% of companies reported using AI in at least one business function, showing gradual mainstreaming. However, only 31% leverage AI across multiple functions and reported financial impacts remain limited for most. 

Generative AI may spur investment, but broader barriers may constrain adoption. Challenges include identifying high-value use cases beyond the obvious (content, customer service) integrating AI operationally, getting buy-in across silos, and measuring impact.

To drive greater adoption and impact, organizations must take a programmatic approach AND a human cost and augmentation approach. This includes auditing data and tech infrastructure, developing organization-wide AI literacy, implementing MLOps, and creating cross-functional AI enabled product teams. 

While generative AI has the spotlight and brings excitement, harnessing the full spectrum of AI technologies in an enterprise context remains difficult. Patience and systemic thinking are vital.

The explosion in Generative AI is clearly shaking things up, but leading organizations should continue to combine emerging innovations with robust strategies, improving capabilities, and value focus to maximize results - while minimizing risks. 

Developing an AI-Ready Culture

The technical challenges of AI get a lot of attention - data infrastructure, model development, MLOps. However, culture and mindset are equally important for AI success and safety.

Companies need to nurture understanding, literacy and trust across the workforce. Access to the best training- not just the tools, but how AI can support roles - helps employees anticipate how AI will impact their roles and prepares them to use AI tools n their everyday work. 

Giving workers opportunities to voice concerns and ideas about AI systems will be important - making your company AI culture and policy visible and undertandable will be key here - for retaining staff, hiring and regulatory. (AI policy as the new privacy link at the bottom of your website and comms) Distributed governance can give diverse stakeholders input on AI priorities and policies within organisations.

"Our biggest challenge is balancing the obvious cost savings with AI and employees wondering about their job security"

 Head of services, FTSE 100 company

Leaders can play a crucial role in cultivating an openness to AI-driven change. clearly articulating the human vision (benefit and challenges) behind AI efforts is a start - effective leadership will also need to model agile learning and growth mentalities as organizations evolve.

Adopting Responsible AI Practices

With advanced AI generative systems, ethical risks around bias, misinformation, can be challenges to adoption. Companies will need safeguards - for reputational, financial and cultural reasons.

Responsible AI begins with understanding what goes into AI systems and these should be tested continuously for harmful outputs - human in the loop and human reinforcement during training can help somewhat with these. Policies and controls should limit overly-risky applications. And transparency helps stakeholders interpret AI reasoning and build appropriate trust.

Companies using private LLMs trained on their own data is a route to consider, though making unstructured data machine readable and executable are not without technical and cost challenges.

Of course, no framework or AI action plan is perfect. But organizations who invest early and continuously in responsible AI practices will better navigate challenges. 

Focusing on Value - And Valuing Focus

The most successful businesses will hone in on use cases that generate clear strategic value - enhanced product or service offerings, improved customer operations, and transformed decision-making with attributable ROI. These can start small with tight scoping, applying test and learn methodologies before scaling - or discontinuing

Organisations that spread their AI investments thinly across too many low-impact pilots and proofs-of-concept will struggle. This approach may fail to produce material benefits or build sustainable capabilities, which can lead to 'we tried that, it didn't work' syndrome.

AI leadership requires focus - picking targets with care then doubling down on what works. It also takes integration with business priorities and processes. When AI is isolated as a pure technology play (without consideration of it's impact across organisations) it's harder to make impactful.

The state of AI right now is full of potential - still to be fully realized. Turning possibility into practical advantage takes foresight, responsibility and focus. The businesses that are leaning in, testing, and adopting new ways to incorporate AI in their strongest use cases are more likely to gain a competitive advantage.

To download the full report from Mckinsey head over here

---

Roy Murphy is co-founder of G3NR8 writes about generative AI, exponential technology and how brands can keep competitive in a rapidly shifting world.

G3NR8 is the AI & Web3 creative consultancy. Helping clients navigate and implement exponential technologies to make and save money.