VR and AI are more than just a pair of catchy acronyms. When the two collide, the resulting ethical fallout could even have HAL’s head in a spin.
Let’s step back and look at these two technologies separately. Virtual reality has surged in popularity over the last couple of years, with the UK Government recently pledging £33 million to the development of immersive technologies.
This leap of faith is no surprise as immersive technologies (including virtual and augmented reality) have the potential to safely train the next generation of surgeons and pilots, democratise education with virtual lessons, revolutionise the travel and leisure industries with entirely new experiences, and bring live concerts into your living room.
Professor Thomas Metzinger, a fellow at the Gutenberg Research College at the Johannes Gutenberg University in Mainz, said: “Virtual reality will have many positive results in, for example, psychotherapy in areas such as the treatment of anxiety. A lot of good will come out of the development of VR.”
For example, Huckletree West members Mindwave Ventures have developed a virtual reality experience for safe use in OCD exposure therapy, in order to treat young people with a phobia of germs.
However, the risks surrounding immersive technologies are largely unknown, which prompted Professor Metzinger and his colleague Michael Madary to publish a series of recommendations on the ethical design and implementation of virtual reality. The report states: “There are several open empirical questions that should be urgently addressed in a beneficent research environment in order to mitigate risks and raise awareness for users of VR in the general public. More research is needed.”
When it comes to AI, the advantages of such systems to optimise our healthcare services, financial sector and retail industry are apparent – and these are just a small handful of beneficial examples. A recent PwC report states: “With a market projected to reach $70 billion by 2020, artificial intelligence is poised to have a transformative effect on consumer, enterprise, and government markets around the world.”
However, from the House of Lords to Elon Musk, many are also eager to understand the risks associated with artificial intelligence. Metzinger recently penned a paper “Towards a Global Artificial Intelligence Charter”, which highlights the five most pressing issues in AI, with practical recommendations and a call for the EU to immediately begin working towards such a charter.
And, while Musk believes artificial intelligence could help trigger the next world war, Mark Zuckerberg claimed such views were “pretty irresponsible”. Responding to Zuckerberg, Musk said his fellow tech billionaire’s understanding of the threat posed by artificial intelligence “is limited”.
However, if we rise above the headlines and Twitter spats, one thing is clear: the true implications of VR or AI on society are far from understood.
Pairing VR and AI
The combination of VR and AI offers many opportunities to many industries. Brands can offer a new deluge of experiences that respond in a more human way – it’s a whole new level of engagement.
The retail sector is introducing many new VR- and AI-based experiences to shoppers. AI bots are predicted to account for 85% of all customer service interactions by 2020, for example, and Adidas has even partnered with Microsoft Kinect to build a body scanner that sits in their physical stores so you can virtually try on clothes.
Facebook is leading the charge in the VR social media space to offer a more compelling experience where 360-degree photos and videos are coupled with the company’s Oculus VR headsets. There are also rumours that an AI assistant for the Oculus Rift system is currently under development.
But what would an AI/VR social media network look like? Well, no one knows. But if you could slap on a headset and simply meet your friends in a virtual pub without leaving the house, would you? If that “pub” was filled with a mixture of AI-based and human avatars, would you know the difference? Or even care?
If we couple VR and AI then the tug-of-war between the pros and cons of such nascent tech gets more complicated. Professor Metzinger added: “When it comes to the confluence of AI and VR, we need to take things more slowly. We need rules, regulations and to be more cautious – but I do not think Silicon Valley cares so much [about that].”
This is where the cracks start to show.
We have not yet been able to interpret the long-term physical and mental consequences of advancing AI or VR technologies – let alone what will happen when the two work together.
A recent Artificial Intelligence: Opportunities and Risk Policy Paper from the Effective Altruism Foundation states: “Several factors currently indicate that these [artificial intelligence] trends are profoundly changing our social behaviour, attention spans, and childhood development. These effects may be amplified by the increasing use of virtual reality technology, which is already available to consumers.”
“As these become increasingly detailed and realistic, they may blur the user’s boundaries between reality and simulation, thereby invading deeper into our everyday experience. The consequences of more regular immersion in virtual realities—including experiences like body-transfer illusions, in which subjective awareness is temporarily projected into a virtual avatar – should receive greater attention.”
This is an important point, as Metzinger explains: “If we are in a virtual reality world, we do not know who an avatar we are interacting with really is. It could be an advanced AI system that is disguised as human – and that is both exciting and threatening.”
This technology is already starting to surface, with Google’s Duplex robot assistant using such a natural speech pattern when users try to book an appointment over the phone, that you may not realise you’re talking to a machine.
Shows such as Westworld have already raised uncomfortable questions about the morality of combining advanced AI, VR and video games. While we’re a long way off from Westworld-esque androids, where we draw the line between what is real and what is not will only get more complicated.
So, when is immersive tech too immersive? “When it damages the self-model in the human user’s brain,” according to Metzinger, and when our moral compass starts to skew so that the consequences of our actions in a virtual world damage others.
Whether by “others” we mean a human or robot avatar is another ethical dilemma.
Because, while today’s androids aren’t capable of dreaming of electric sheep just yet, Silicon Valley and world leaders must tread carefully as AI/VR crossovers become more prevalent.
To quote American author Kurt Vonnegut:
“We are what we pretend to be, so we must be careful about what we pretend to be.”
Startup working in mixed realities? Pushing the boundaries of AR/VR/AI? Apply now to join Huckletree West’s resident community of pioneering immersive tech businesses.