While building and testing new products used to require significant expense across multiple areas, AI agents cut the cost of experimentation to almost zero
Enterprises choosing between open-source and commercial AI models must carefully weigh risks and benefits, considering security, IP ownership and liability
AI systems require vastly more memory than consumer devices and supply chains are prioritising the needs of AI infrastructure rather than consumer electronics
AI is becoming an ‘autonomous worker’, capable of handling the full customer support cycle from beginning to end. What should business expect from this change?
AI agents are customer-facing, acting on real-time data, often without a human in the loop. Most organisations don’t realise what that means for their business
Activism itself is not misconduct but organisations should proactively manage actors who amplify existing grievances through narrative shaping or selective leaks
If an agent can publish persuasive content, recruit other agents or influence decisions, we need to know what it is and what it has been authorised to do