Modern businesses are rapidly evolving as Artificial Intelligence (AI) moves from pilot projects to large-scale implementations. This ‘AI at scale’ revolution is transforming industries, driving innovation, and enhancing operational efficiencies with AI being deployed at scale across various business functions.
The transition from PoCs to large-scale AI marks a pivotal moment for businesses. Initial projects showcase potential but usually fall short of full-scale impact. It is often observed when AI is implemented organization-wide, it greatly enhances efficiency, decision-making, and competitive advantage. For example, a European electronics retailer uses AI to predict online shopper conversions and prompt interventions during a browsing session. Similarly, a European publication leverages Gen AI to summarize reader sentiment from comments and reviews to score and improve their articles.
However, according to the “Global AI Adoption Index” report, European businesses have been reluctant to adopt AI due to regulatory and safety concerns. The European Court of Auditors confirmed that EU AI investments lag behind global leaders. To address this, the EU implemented the AI Act, which mandates specific AI uses and reduces administrative and financial burdens, and places compliance responsibility on AI solution providers, recognizing that many businesses may not have the AI expertise. The AI Innovation Package and the Coordinated Plan on AI also aim to boost AI investment across the EU.
As businesses in Europe find it lucrative to increasingly adopt AI technologies at scale, they can expect significant advantages in internal efficiency and innovation without a large regulatory risk.
Internal efficiency
Scaling AI improves internal efficiencies by automating low-risk tasks, allowing more focus on strategic activities. For instance, AI can manage customer service inquiries, complex supply chain logistics or credit risk scoring. A leading European bank uses AI to analyze client documents and conduct credit risk assessments, even in “low resource” languages (languages with low digital data to learn from). Mortgage officers retain decision-making authority, the AI solution an untiring assistant.
Innovation and success
Scaling AI fosters innovation and enables uncovering of insights and making smarter decisions. For example, a European consumer goods firm uses AI to boost digital engagement by analyzing search terms and product descriptions in comparison to competition on third-party marketplaces. This helps them highlight key features and use apt keywords in the description. This insight is now also informing on product development.
Successful examples of AI at scale
To grasp AI’s impact at scale, let’s explore its successful applications across various business functions:
Customer service: A large European bank’s contact center uses an award-winning AI solution to assist agents by suggesting resolutions, re-routing calls, predicting intentions, and recommending response language using Gen AI. Integrated with the contact center application, the solution reduces change management complexity. Now widely used, it has improved customer satisfaction scores. And thereby reduced churn.
Human resources: Gen AI usage in HR is risky because it propagates inherent bias existing in historical selections. However, with guardrails that avoid PII data feed at the entry point itself (avoiding post codes, names, genders), and style-masking through summarizing rather than the original text being fed, these risks are being mitigated. The bias removal and improved accuracy are leading to usage at scale.
Supply chain management: A leading European CPG company is implementing an advanced AI forecasting system that autonomously manages demand forecasting. It automatically groups products into demand-forecasting-units (DFUs), selects optimal algorithms, self-corrects based on recent errors, and also uses Gen AI chat to explain exceptional forecast movements for over 20,000 store-SKU combinations, all without manual customization.
Marketing: Your website is often considered your digital billboard. Large apparel retailers are leveraging Gen AI to streamline web content creation by drafting content from various data sources, such as product master data, vendor information, PDFs/PPTs, and marketing digital asset libraries, thereby reducing content creation timelines. This Gen AI-generated content reduces creation timelines and undergoes human review and approval before publication.
Overcoming hurdles and key insights
While the benefits of AI at scale are compelling, businesses must also address several challenges:
Ethical considerations: AI systems must be designed and deployed ethically to avoid biases and ensure fairness as we saw in the HR example above. The AI Act in Europe will further assist companies in adopting AI ethically by providing a regulatory framework.
Technological infrastructure: Businesses must invest in the necessary technology to handle the demands of large-scale AI applications. Responsible AI is not only about regulations but also costs. For example, if one can automatically generate marketing content, it does not mean one should do that every day.
Change management: Effective change management strategies are essential to ensure a smooth transition and gain employee buy-in. When a contact center agent gets a prompt from an AI tool instead of a previous rules-based one, this embedding reduces any pushback.
The ‘AI at scale’ revolution is not just a technological advancement; it is a strategic imperative. While Gen AI is expanding AI possibilities, it is also bringing risks such as black box biases, prompt injections, jailbreaks, and potential runaway infrastructure costs. Without fail safes, a Gen AI system having an outage similar to a Crowd-strike event could be much more damaging. Responsible AI is about costs, employees, society and regulations. Think and thrive responsibly!
We’ve featured the best AI chatbot for business.
This article was produced as part of TechRadarPro’s Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro