Mark Thirlwell, Global Digital Director at BSI.
getty
In many ways, the playroom of today looks little different to that of a generation ago. Children still play with toy cars, trains, dinosaurs or dolls. They build Lego, compete at Monopoly or play Snakes and Ladders.
Yet a closer examination makes clear what’s changed. It’s not just about play but how all of us live. New BSI research has found that half of UK children have access to an AI-enabled toy or learning device, tallying with US research earlier this year. That could be a robot that answers questions, a tablet with voice-assist functionality, or a device intended to teach them numeracy. This is only the start; many major toy companies are investing in AI, including Mattel, Hasbro and Lego. Equally, devices such as Amazon’s Kindle Fire tablets have AI-powered features.
We are on the cusp of a world in which everything we do is enabled by AI, from play to healthcare to shopping and home security. It’s an opportunity for society to progress, for new opportunities to emerge and for more time to be freed up for leisure or spending time with loved ones. But as AI becomes omnipresent in the products we use—from toys to toothbrushes, medical equipment to musical instruments—the question becomes: how can we know whether to trust it?
At present, there’s no clear answer. The EU is pursuing a regulatory approach via the EU AI Act, although progress is slow, while other countries are prioritizing innovation. Alongside this, ISO/IEC 42001 offers the world’s first management system for AI, encouraging organizations to meet its provisions in terms of governance, ethics and responsible use.
Beyond this, what is needed is a shift in how AI development is undertaken, towards safety by design. Ensuring that when AI is integrated into products and devices, they only reach the market after checks have been carried out to confirm they are safe for use. And not just physically safe, but unlikely to cause psychological harm, privacy or data issues, the spread of disinformation, or any other negative effect.
This would build on past success to instill both security and privacy by design in innovation. Indeed, we have long had safety-by-design for physical products, from cars and airplanes to toys. BSI is 125 this year and has helped develop modern consumer safety frameworks in areas from seatbelts and buggies to button batteries. Today, for example, we do not drive off without confirming a child seat’s safety, we rely on standards that are often testable to be confident against unacceptable risk.
As life moved online, it became clear that information security was critical, and security by design emerged. A British standard from 1995 formed the basis for ISO/IEC 27001, covering information security management, cybersecurity and privacy protection. Today, globally, over a million organizations are certified to this approach, which requires organizations to embed security considerations into how systems are conceived, built, operated and improved, rather than being bolted on later.
Of course, this doesn’t prevent all hacks or data leaks, but it does ensure businesses have the protections in place to help data remain confidential, accurate and accessible only to authorized individuals. Without security by design, businesses would be more vulnerable to breaches, cyberattacks or operational disruption, and all of us would face far greater risks of identity theft, financial fraud, privacy loss and service disruption.
The other concept that has emerged has been privacy by design, which has become integral as things like smart doorbells, Alexa and other connected technology. Again, standards have been critical here. ISO 31700, launched in 2023, operationalized a privacy-first approach for consumer goods and services, filling a practical gap left by regulations like GDPR and defining 30 high-level requirements from proactive risk assessments to lifecycle privacy controls and default settings. It gives businesses clear, implementable guidance. Organizations now embed privacy into product architecture and operational practice. As a voluntary but authoritative framework, it has elevated privacy from afterthought to design principle, encouraging responsible innovation.
As AI becomes increasingly embedded, we need security and privacy by design in the conversation and to bolster it with safety by design. Standards can and must be a central part of the conversation. They have always underpinned safety for physical products, and should now underpin digital, connected and AI-driven aspects of safety.
BSI’s research also explored parental fears about AI-enabled toy safety, security and psychological impact. Amid ongoing debate over social media age limits, there’s a gap between how quickly technology is being adopted and the guidance available to families. For example, 91% of UK parents say a recognized safety certification or mark for AI toys would be important, with almost a third describing it as essential. 83% believe manufacturers should comply with established standards or codes of conduct.
Currently, we lack the frameworks to support safe, transparent, age-appropriate use. Clearly, creating them is not simple. Technology moves at pace, and existing safety regimes were built for static physical products—tested at the point they leave the production line. AI-enabled toys are connected, adaptive and update over time. That makes building safety into them more complex, but also urgent for society and governments to grapple with.
Two other striking figures: 66% say we need to learn lessons from young people’s use of social media and apply them to AI technology. Nearly half of parents think their child would be better off in a world without access to AI.
Yet AI is here, in our schools, playrooms and our shops. We can’t switch it off, but we can put energy into ensuring it is designed in our best interests and into establishing the frameworks that help parents and all of us trust our choices. Safety by design should become the baseline.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。