Artificial intelligence (AI) is nothing new. According to Lloyds of London’s 2019 AI Report it has been a significant presence on and off for over 60 years. But now there’s a rapid escalation of AI and people are starting to think of the benefits as well as the disadvantages and technological complexities that come with it.
In my industry (media) I’ve seen AI being increasingly used in the newsroom. The Times newspaper, for instance, has just introduced an AI powered ‘digital butler’ to ensure that they are delivering editorial content in the right place, format and time.
It’s astonishing what James, The Times’ new tool can do. It can apparently analyse user habits to the degree that it can send newsletters with or without pictures (depending on preferences which it has calculated) and knows exactly what time the reader is likely to open the content and read it.
It’s taken time to develop it though – according to a report on journalism.co.uk, The Times collaborated with a software company two years ago and is only now seeing the benefits. But it’s been worth the wait as 70% of subscribers in the trial interacted with James either by opening or clicking on the newsletter. A lot of companies would kill for that kind of an open rate!
For those of you not quite up to speed, the Merriam-Webster dictionary defines AI as ‘the capability of a machine to imitate intelligent human behavior’. Others have defined it the ‘science of creating intelligent machines capable of performing real-time tasks at a level of a human expert’.
According to the Lloyd’s’ report, AI can help with:
- Processing large amounts of data at higher speed and greater capacity than people could
- Learning from example data sets. It solves new problems from this data too.
- Recognising objects and their association to a situation
- Inferring the future state of an object or situation
- Identifying an optimal decision based on past, present and inferred future states.
AI is being used in a number of different sectors. It’s being used to help power driverless cars, in helping telecommunications companies with their retention strategies and e-commerce businesses to improve their interaction with customers who engage with them online.
It’s also been used to help in farming and medicine. Farmers can make better decisions on planting the right seeds and fertilisers while medical experts have used AI to help diagnose diseases, develop drugs and personalizing treatment.
AI is of course also increasingly used in banking and finance. AI is being used to recommend investments based on client’s risk profiles and it’s also being used to help reduce and detect credit card fraud.
Lloyd’s of London says in its report that it has used AI software to build a system that is now starting to provide fast and accurate knowledge to the Lloyd’s International Trading Advice (LITA), which answers time-sensitive trading questions from the Lloyd’s market.
So, what about jobs? This is addressed in Lloyd’s of London’s report too. Because with great innovation often comes a great number of jobs lost. Think about all of our past industrial revolutions where technological advances have kept many at a standstill – unable to keep up with the momentum and drive of technology.
AI is bound to have an impact on jobs too. Lloyd’s says this is likely to affect those in the lowliest skilled jobs. But on the other side of the spectrum, the introduction of AI is set to create a demand for people that are skilled.
It points out: ‘The lack of suitably-skilled experts in AI is one of the more significant challenges for its greater adoption’. According to a Deloitte report which looks at the impact of technology on jobs in the UK, around 800,000 low-skill, routine based jobs have dropped disappeared between 2000 and 2015.
While the report says that AI and automation is likely to only affect manufacturing and agricultural industries it’s likely to have an impact in the insurance and legal professions especially in positions that rely on sourcing documents that are already stored digitally.
But there’s still a skill shortage when it comes to the upper end of the skill spectrum. So, what’s really killing the wider adoption of AI is the lack of PhD and Mastered candidates that can help to interpret what AI software spits out after it masters and analyses the mountain of data we give it.
With nearly every generation it is said that jobs for the young, school generation have yet to be created. But if AI will be embraced as much as we think it will we will definitely need more data scientists that can help to drive this to its full potential.
In the UK, for example, we’ll need about 300 industry-sponsored Masters-level places per year and 200 PhD-level places per year offered by universities with government support according to a report by Hall and Pesenti.
So, what’s the solution to the potential job losses? Well, some countries have already toyed with the idea of introducing a universal basic income (UBI) for all its citizens and so far, the idea has been trialed in Kenya, Scotland, the United States, Canada and Finland.
The idea of a UBI has even been mooted here in the UK. A basic income pilote is being considered in Glasgow and just this month a report by Professor Guy Standing from the SOAS University says Britain should introduce a weekly UBI of £48 a week. It would cost about £150 billion a year and Standing says this can be achieved through scrapping more than 1,000 tax reliefs.
UBI has been criticized by many – some of whom describe it as ‘free money’ for everybody. While the Labour Party has said it would look into a it, Conservative deputy chairman Helen Whately said it would be a ‘kick in the teeth to hardworking taxpayers’ according to the BBC.
She said: “It would mean benefit payments for everyone, from Premier League footballers to investment bankers and even prisoners, costing billions, while hammering ordinary workers in the pocket by scrapping the tax-free income allowance.”
The Lloyd’s of London report also suggests other income protection schemes such as AI tax arrangements and that better protection insurance solutions should be introduced that could help to stem the impact of AI.
Then there is the potential AI has in terms of helping those with malicious intentions to attack our security and defense systems. It may sound a bit like some kind of science fiction or Terminator movie, but it’s been suggested that autonomous drones could be used as weapons. We’d have all manner of intrusions into our manufacturing and services sectors and businesses could be at bigger risk of more sophisticated cyber-attacks.
AI, says the report, would have an impact on our digital security, physical security (think crashes between two AI-powered cars) and our political security – think invasion of privacy as well as manipulations of election results to name but a few scenarios.
For now, the most likely and first danger that AI will present is a threat to our job security. Hundreds upon thousands of jobs will be lost in a short few years if nothing is done at government or company level to address this potential threat.
Millions of pounds need to be invested in future learning. And that doesn’t mean only investment for the generation that are at school level but for the thousands of people that are in low-skilled jobs that are likely to lose out.
It’s likely that the benefits AI will bring will far outstrip the downsides. It will help to create jobs and raise us all up to never seen before levels. It will help us to decipher and interpret mountains of data that would usually take us hours, months or even years to wade through. Who knows what sorts of benefits we will reap from that. A cure to cancer? Or will AI super computers provide a solution to time travel?
Will AI help us see into the future? To an extent it already does. As I mentioned earlier there are newspapers that have already created digital butlers that are catering to our every need by giving us exactly what we want to read and presenting it to us in a way that’s pleasing to our eyes.
Isn’t it time that we, as an industry, embrace AI – particularly if it’s giving us exactly what we want? No more time wasting. No more second guesses. No more sending out multiple campaigns and emails hoping that one of them will resonate with a few hundred people.
The insurance and financial services sector are already viewed by many consumers as a sector that can’t be trusted. Perhaps AI will help the industry gain back that trust by giving customers exactly what they want and not what we think they want.