Imagine a business world where employees are faster and more productive – where they can make smarter decisions and have the time to focus on strategy and being creative. This is all a near-reality with continued breakthroughs in artificial intelligence (AI) capabilities. AI is at the tipping point of becoming the next great technological disruptor.
Improvements in computing power, the advent of big data and breakthroughs in machine learning have created the ideal environment for AI to flourish and augment human potential. At the very least, AI will transform our economies, reshape consumer expectations, and increase the speed and scale of business. At it’s limit, it will help us understand humanity and human intelligence better.
As Salesforce’s Chief Scientist, I lead Salesforce’s research efforts, with a focus on bringing AI solutions to all businesses. My background is in AI, my passion is AI, and particularly in deep learning – the subfield of AI that’s focused on learning from raw data such as natural language or visual inputs.
I was recently in Australia talking about ‘The Edge of the Possible’ and, in between photographing kangaroos and climbing the Harbour Bridge, had some great discussions about AI – including where it stands today, the potential for the future, and what this all means for businesses.
As in past periods of technological disruption, AI will unleash new levels of productivity – a new International Data Corporation (IDC) report predicts that increased productivity will boost global business revenues by $121 billion from the beginning of 2017 to the end of 2021, contributing to a total $1.1 trillion revenue boost as a result of AI associated with CRM.
Australian businesses are expected to see a $29 billion bottom-line boost in the five years to 2021 – a result of increased revenue, higher productivity and decreased expenses.
The report also predicts that more employment will be generated by AI-powered CRM than that lost to automation, with 800,000 net-new jobs expected globally.
AI is set to transform our personal and professional lives, and determine who the next set of winners and losers in business will be.
All of this is because AI will revolutionise every aspect of business. For example, customer service can deliver proactive help, resolving issues before they happen. Sales people will be able to predict which opportunities are most likely to close and receive recommendations on how to close each deal. Marketing can create predictive customer journeys and personalise experiences like never before. IT can embed intelligence everywhere and create smarter apps for employees and customers.
Yet, as AI advances, it’s important that all companies can access and benefit from the technology. As it stands, the technical complexity and resources required to deliver AI are out of reach for most companies. This is something Salesforce is committed to changing with Einstein, which embeds advanced AI capabilities in all parts of the Salesforce Platform. In many cases we also provide standalone APIs, offering all of our customers access to AI.
There are three ingredients for businesses to be successful in the realm of AI: access to data, good algorithms and smart workflow integration.
The first step is having access to your data. That data has to be in a structured, usable form, and you need to have a clear idea of the desired outcome. If you don’t have this, there’s no point proceeding.
Secondly, you need strong algorithms. Fortunately, there are a lot of great vendors, like Salesforce, that can test default algorithms on your behalf. Then, if you have a research group, you can push these algorithms even further, creating new capabilities.
But the most important part is smart workflow integration. Your AI features need to be incorporated in a way that makes people feel empowered, not like they're being replaced. You need to first understand the existing workflows you have within your business, and infuse AI in the right places, so it’s a seamless experience. The best workflow integrations continuously improve AI training.
In this video, Richard Socher explains why AI will improve the way we work and how to embrace the technology to get ahead:
As different parts of the puzzle are cracked, AI, as we know it, continues to evolve and change. We’ve had breakthroughs, like solving speech recognition and identifying objects in images. There have been huge advancements, largely thanks to deep learning. But we haven’t solved the larger AI problem – general, or human-like, intelligence is still far away.
The most exciting space at the moment is natural language understanding. The IDC research found that 40% of companies plan to adopt AI within two years, and forecasts that 75% of enterprise and ISV development will include AI or machine-learning functionality in at least one application. Of those companies planning to adopt AI, 30% are focused on voice and speech recognition.
But natural language understanding is also where we have the most work to do. Yes, we can ask Siri a series of questions using our smartphones. But there’s still a lot of research to be done to truly understand language and dialogue – how we communicate, how we think – and have this represented in a computer in a connected way.
And if you think about ways customers interact with businesses, it’s through language: voice, email, text message, social media, etc. So the more we feed this visual and textual information into the computing machine, the more it understands the way our customers talk and interact – this will become extremely powerful in the future.
Yes, there are existing capabilities that already work well in this field, such as search technologies, recommendation engines and ecommerce websites – all multi-billion dollar industries. But the next step is making state of the art technologies, such as these, widely accessible to the masses, ensuring they are useful for all businesses.
Finally, as human intelligence merges more with artificial intelligence, we’ll no longer have to do repetitive work. We’ll be able to focus our energy on the creative tasks that can’t be automated – the tasks that make us uniquely human.
AI is set to change everything about customer experience and to push customer expectations higher than ever, so top IT teams are hyperfocused on transforming tech to provide the best possible experiences. Find out how they are doing it in our second annual State of IT report.