As the weather heats up, the tech leaders come out of hibernation to tell us what will be big this year and beyond. While it comes as little surprise that folding phones are suddenly off the agenda, the rise of artificial intelligence and chatbots continue to dominate as Google, Microsoft and Facebook talk shop.
Facebook F8 Developer Conference (April 30 – 1 May)
Mark Zuckerberg somehow remains in charge at Facebook despite making off-colour comments about user privacy, as if this is all still a big game to him. Even so, the news from the F8 event that Messenger is getting a major overhaul this year shows that he understands the need for an AI and chatbot-led future.
Facebook Messenger will benefit from end-to-end privacy, as part of the company’s new ‘The future is private’ ethos. That was among the standout news alongside new virtual reality headsets, the choice to see what your contacts are doing on other social media, as Instagram, WhatsApp and Messenger are integrated to some extent.
That and Facebook Dating will all create new opportunities for improved chatbots and help provide businesses with more features like appointment booking via bots in Messenger (see image) and lead-generation tools to help improve sales and advertising value.
The privacy issue will be powered by Facebook AI tools to make it real time, along with artificial intelligence tools that use supervised self-learning to spot and remove hate speech more efficiently. Supervised Self-Learning reduces the amount of training data the company requires by around ten times, depending on the product and is key to getting to the AI to learn to be smart with a minimum of human input.
For developers using third-party tools to push chatbots into Messenger, there may be some disruption getting bots to operate with the redesigned apps, but Facebook’s reliance on partners for services should see the company trying to avoid upsetting the developer base.
Microsoft Build (May 6-8)
It looks like AI will soon be at the core of everything Microsoft does and its millions of customers use. For example, announced at the company’s Build event was an AI feature for Word Online that adds Grammarly-type language and style suggestions to improve the quality and readability of a document. That’s a small addition in the big scheme of our AI future, but it could improve the writing of millions of users and the readability of billions of pages thanks to AI’s ability to recognise context and nuance.
Wearing my author’s hat, while the creative side of content generation has a long way to go in terms of AI, many of my writer friends swear by tools that assist with good grammar. And it’s interesting that you can date a book by how it reports the exchange of text messages. Those full of spelling and grammar mistakes in the chatbar are set in the 1990s and 2000s. Today, our messages would delight figures like Professor Huddleston.
Microsoft is putting plenty of effort behind speech, even as Cortana struggles for wider acceptance. However, developers are keen to show the way, with tracks on how to design for speech and how to inject personality into bots, ‘enhances your bot’s conversational capabilities by handling small talk, in line with a chosen personality. Choose a personality that aligns with your brand’s voice, by choosing from available default personas.’
Another, rather more dramatic example is BMW, a major Microsoft partner, showing off its Azure cloud-powered in-car personal assistant that uses Microsoft’s Bot Framework Virtual Assistant Solution Accelerator’s Cognitive functions to add smarts for road users. That’s alongside more practical talks and presentations such as ‘How to prevent biased datasets when training AI models’ and ‘5 industries that are getting disrupted by Computer Vision on Cloud and on Edge’.
Google I/O (May 7-9)
Google’s Duplex was the star of 2018’s I/O Event and in 2019 the AI-powered voice assistant comes to the web, providing audio help to any website with no server-side changes needed. It can provide machine-to-machine functions, help users fill out forms and much more, bringing AI services to a new level in the traditional browser.
Within Android, Google showed off new AI-powered features for Search and Lens, with the company highlighting that ‘people have already asked Google Lens more than a billion questions about things they see. Lens taps into machine learning (ML), computer vision and tens of billions of facts in the Knowledge Graph to answer these questions. Now, we’re evolving Lens to provide more visual answers to visual questions.’
New Lens tools include bill splitting, translating menu items, narrating, all faster than ever before. Not only does this make the camera more useful in a phone, beyond selfies, but it links to other apps and services to help deliver the best result, faster. While the new Assistant can multitask across apps and maintain a running conversation without the need to say ‘Hey Google’ all the time. At quite a rate of knots, AI and bots are stitching themselves into the fabric of what we do and how we all interact with technology, making future chatbots and other assistants seem like part of the family.
Touting ‘AI for everyone’ Google was also keen to promote the privacy and security aspects of its technology. The company promises products and services that are ‘built on a foundation of user trust and privacy.’ And, while most of look at technology as a cool or functional gadget, CEO Sundar Pichai promised to do some good with all this AI power, ‘Over 1 billion people have a disability. We believe technology can help us be more inclusive, and AI is providing us with new tools to dramatically improve experiences for people with disabilities.’
This resonated with me and in my own chatbot building work, I hope to see more utility from the technology for the visually impaired via text-to-speech.
One of the first examples is subtitles for any video using AI to figure out what’s being said and live transcription of phone calls, all ideal for those in a noisy environment or the hard of hearing. There’s plenty of tracks to catch up on from the event, with Actions for Assistant the way forward for useful user engagement, quality conversations and insights.
All of these tools need to be built without bias and to cater to a specific or wide audience without causing upset or offence. Google’s boss was keen to point out just how easy it is to go wrong. While any business can build these tools, using Google’s or other services they do need to be very careful about avoiding bias and poor quality data that could lead to unhappy customers.
Big Tech Helps Build Small Dreams
While these tech giants might throw their massive AI budgets into the sky and expect us to go ‘wow!’ at all the pretty resulting fireworks, it is still largely the smaller companies and end-user businesses that will turn these features into the magic that drives experiences and revenue.
For example, there are now thousands of games available that use Amazon Alexa’s voice features, the most popular augmented reality apps all come from innovative developers and developers will find plenty of uses for all these new features that the big brands have never thought of.
Across all these vendors and the massive number of services, the clear push is toward making apps and features better integrated, while improving user privacy. The move to voice is the big push forward this year, as people move to screenless or Watch-type devices that require audio interaction.
On the hardware front, with the drying up of premium device sales and the extension of upgrade cycles, Google and Facebook were both focused on lower-cost hardware. Among the slew of developer news was new cheaper hardware like the Google Pixel 3A phone and lower-cost VR gear from Oculus. Will Apple do the decent thing next and look at a genuine mid-range iPhone?
Note, for Apple fans, its worldwide developer conference (WWDC) runs from 3-7 June and will likely riff on similar services, privacy-focused smart tech announcements, all built on AI services. iMessage is getting a fresh update and will likely renew the call for businesses to run chatbots for all sorts of use cases.
And it is not just the big tech names that are trying to push AI and bots. The likes of Oracle is running bot and virtual assistant events over the year to help businesses. Lenovo’s upcoming Accelerate event has a major focus on AR/VR, AI and IoT to show enterprise customers the value of these services as part of doing business.
Whatever your organisation size and market, soon every company will be using AI and bot or virtual technology to act smarter or more efficiently. No business can afford to be left behind, and thanks to the as-a-service nature, most companies can build these tools without dedicated developers or at great expense.