DataIQ (DIQ): Generative AI has been the most discussed topic in recent months. How has News UK been approaching the topic and the new tools?
Will Sach and Pedro Cosa (WS&PC): News UK is home to a raft of multi-media brands which together reach over 40 million people. Our focus on performance is part of a set of values which puts data and technology at the heart of our business and decision making. As our business continues to evolve, so does our approach to creating value and delivering sustainable business growth.
By the same token, the use and benefits of AI are not new to News UK. Over the past five years, AI has supported several use cases that have driven the performance of the newsroom, marketing and commercial teams.
The emergence of generative AI tools offers opportunities to further augment the generation and commercialisation of our content and achieve wider cost savings by improving operational effectiveness. However, it also poses risks for our business, such as how we protect the uniqueness of our original content and remain a trusted news source.
To manage the deployment of generative AI, we have considered the cost of implementation, controlled versus uncontrolled deployment of AI tools, regulatory obligations and the risks to our people driven by the excitement or doom-and-gloom scenarios associated with AI more broadly. We then set out on how to manage the explosion of interest across our organisation with the right balance of control, so that we could seize new opportunities quickly and without stifling innovation.
DIQ: What do you think is the most exciting aspect of generative AI, and are there any aspects that data leaders should approach with caution?
WS&PC: The art of the possible has just changed. Our ability to augment the speed of content generation, improve its overall quality and commercialisation across text, audio and visual formats is exciting. For instance, there are new efficiencies we can drive in converting audio to text. The ability to summarise radio match reports into articles using original source content will make our newsrooms far more productive.
We have witnessed an explosion of data usage. For years, we have been talking about democratising the value of data and now through new public instances of generative AI everyone can access, interrogate and create value from data. We now have data at our fingertips. Across the analytics and insights space, we can look at ways to summarise insights more quickly and succinctly to provide actionable recommendations back to the business.
On the other hand, all businesses need to be mindful of the rise of bad actors. Generative AI can be used to generate fake news, propaganda and other forms of disinformation, which can harm the credibility and reputation of professional journalism.
Our intellectual property remains paramount and generative AI may infringe on copyrighted or trademarked material, which could lead to legal issues for publishing businesses.
Finally, it is important to protect free speech and opinion, ensuring that the use of AI tools does not lead to bias or discrimination.
All businesses need to be mindful of the rise of bad actors.
DIQ: As a media business, what do you think AI will be able to bring your team and the wider organisation?
WS&PC: Enhanced content planning and creation – including content research, identifying topics and trends, article summarisation and headline optimisation.
More efficient newsrooms – increasing the automation of time-consuming tasks like content tagging, fact checking, image detection and audio to text conversion.
Improved commercial outcomes – providing more creative ideas and decks in response to client briefs or improved advertising planning by identifying reader segments more clearly for advertisers.
DIQ: On the other hand, what do you think the limitations of AI are for your business objectives?
WS&PC: The phrase “there is an AI for that” is already becoming quite popular, but we need to be mindful of its strengths and its weaknesses. If we take this phrase quite literally there is a danger that we will begin to take our eye off the ball on when and where human involvement should take place. For instance, as a premium publisher, it will never replace journalistic instinct or good judgement in the timing of when to drop a story to deliver maximum value. So, we need to be clear on where to rely on humans over machines and why.
The regulatory risks in using AI cannot be ignored. At a time when privacy regulation is strengthening, we are fast reaching a place where the data we can use is only as good as the consent levels we collect. Even the best AI tools available will become redundant quickly if the right data permissions have not been sought out at the start.
Lastly, choosing the right mix of AI tools will become increasingly more important as will the need to have a thorough understanding of the individual strengths and limitations of each set of tools. This could help avoid making the wrong investment decisions which ultimately should be use-case led.
As a DataIQ member, we have received the opportunity to share our experiences of data with industry peers in a safe environment as part of a wider community.
DIQ: You recently received an AI Innovation award from DataIQ. What sort of AI initiatives have you been implementing and what does the next stage of your AI journey look like?
WS&PC: Broadly speaking, we have focused on three main areas; operational efficiency, business effectiveness and IP protection. Operational efficiency covers things like improving content tagging, editorial content production and internal operational workflows. Business effectiveness looks at enhancing analytics, insights and reporting, non-editorial content generation and augmenting external customer journeys in our contact centres. Lastly, IP protection focuses on ways in which we can ensure fair value for our content, identifying secure technologies which protect us from data leakage and adhere to our News UK policy or AI guardrails.
The next stage is examining how to deliver value from AI at scale by reviewing the necessary infrastructure, investment and training requirements to achieve our business goals.
DIQ: As an active DataIQ member, how has the DataIQ community assisted you and your team with your data journey?
WS&PC: Fundamentally, as a DataIQ member, we have received the opportunity to share our experiences of data with industry peers in a safe environment as part of a wider community. We have taken great benefit from being able to draw from a well of knowledge on topics in which our peers are facing similar challenges like data strategy, leadership and effective data governance.
We have also been given access to great content, enjoyed participating in roundtable discussions and benefitted from the networking opportunities across the fantastic calendar of events run across the year.
DIQ: If you could give any advice to a data team looking to embark on their AI journey in today’s business environment, what would it be?
WS&PC: We would start by saying ride the wave of generative AI and leverage its momentum. It is a great opportunity to put data at the top of the agenda.
If there were ever an opportunity to elevate environmental, social and governance efforts it is now.
Focus on solving business problems which can solve pain points and deliver value quickly. They will help generate further support and the continued investment needed.
Promote innovation with a certain level of control. The key is not to slow things down but make sure things are done properly, managing both risk and reward.
Double efforts in governance and education. People need to know the rules of the game and the tools they can use before they start. The financial, regulatory and reputational risks are real and often there will be unintended consequences driven by improper usage.
Finally, if there were ever an opportunity to elevate environmental, social and governance efforts it is now. We know large language models have the potential to consume vast amounts of energy, so we must use them responsibly and with full understanding of their impact on our environment and our communities. This led News UK to the creation of our final ethical principle: Sustainable.
Both Tom Jackson, director of data technology and Harry Burt, director business innovation and transformation at News UK were consulted by Will and Pedro for their additional input and insights.
To better understand your own AI journey, click here to take the free DataIQ AI Indicator Assessment.
Click here to download the DataIQ advisory report on gen AI.
