What Is The Future?

Introduction

We are at an inflection point in technology. My thesis, now nearly a decade old, might be realised in this coming decade. As I’ve tried to explain where we are, AI is a general-purpose technology like the printing press or the steam engine, which has consumed all the text on the internet, or a tool that will augment us. But I still believe in the idea that we will have digital companions as new partners in the journey of all our lives. But before we get there, we have to state that where we are currently is still close to a digital tool, at least what is in the public domain. 

Technology is a tool, but it's not the end. Have we started to innovate, or are we struggling to do so? In terms of our ability and will to innovate, let’s take a look at:

• What is possible

• What is likely to happen

• What is desirable to have happen

Are we at an inflection point where we spend too much time refining cool, shiny technology with sentiments like “Look at this cool thing, it’s awesome”? Would it not be more useful to shift this mindset and focus instead on the problems that need solving and developing a deep understanding of those? More than just a suggestion, this shift is necessary; without it, we could miss opportunities for remarkable discoveries. It will invoke the exploratory nature of science as expressed by Richard Hamming in his book The Art of Doing Science and Engineering: “If you know what you are doing, you should not be doing it. If you don't know what you are doing in engineering, you should not be doing it.” Richard Hamming The early Greek philosophers Socrates (469–399), Plato (427–347), and Aristotle (384–322) believed anything can be talked about in words. This belief ignored the mystery cults of the time who asserted that some things - the gods, truth, justice, the arts, beauty, and love - could not communicated in words; you could only “experience” them. So it is now: rather than expressing ourselves in words, we have to build, make, test and solve problems for people today and in the future. 

What is Possible and How We’re Holding Back

The World Wide Web's standardisation and decentralisation of computing and data have allowed more and more people to access its power. Over half the world’s population uses it, which has truly changed our lives. Since the world stopped with COVID, there have been endless WhatsApp chats and pub conversations about emergent technologies. Discussions which will fundamentally change how we ‘do computer stuff’. I’m most interested in how culture affects how we assess these new forms of computers and why we appear trapped in a cycle of incrementalism. While prevalent in the tech industry, this phenomenon warrants a deeper understanding and exploration.  While there’s no denying the progress of technology in the last decade, it’s true that our devices - how humans access much of that technology - have not changed significantly. Plenty of small, incremental changes have emerged, but nothing ‘new’ has emerged. Although we have smart speakers, voice UI hasn’t become an essential or dominant interaction beyond selecting playlists and setting cooking timers. The TV has a necessary place in our technology experiences of late, but fundamentals have remained remarkably the same for years, with most of the change coming via the pipes that feed the content. Games consoles could be attractive, but they are still just boxes you plug into a TV, which has remained unchanged since the Atari days. These ‘connected things’- doorbells, cameras, smoke alarms, etc. - have had tiny updates but are essentially the same as they always were. Computing has not changed these ‘one box’ solutions. The essential premise has remained the same for decades: a large screen for productive work paired with a powerful processor, keyboard, and mouse for input. We saw the emergence of the laptop in the mid-eighties and now carry these around in our backpacks; however, the laptop’s design has remained the same for over two decades. The iPhone landed in 2007, and tablets arrived shortly afterwards in 2010. The ‘smartwatch’ had been around for over fifteen years in various forms before Apple started to dominate the market in 2015.  While many new features have since been added to these devices, the focus has been on small iterations and product evolution, minor updates, and refinement. The fundamental changes the consumers have felt over the last decade have come mainly via vast amounts of back-end work, content acquisitions, hardware refinement, software integrations and partnerships, which escape the interest of the everyday consumer. The world of personal computing is stable and mature, but there are signs that it might change. 

Think of the Present

Most of us tend not to think past our current state and find it hard to think about how things could or will be in the future. We tend to think of ourselves as rational, right and proper. We can consider all available options and think our choice is correct. It is nearly impossible to anticipate how society or individuals might change, how our habits will change, or how our tastes and attitudes could develop. Knowing this, can we trust our decisions? We must think further into the future and understand that change is inevitable.  We might now look back at the 1990s and good-humouredly laugh at the design of our homes, the vast Walkmans with those beautiful large buttons and the click sound they made, the funny-looking VCRs, the clothing, and the music. These were the tangible and available technologies and the habits and tastes of the moment, but today, they feel strangely outdated. I’m confident the next generation will look back and have the same attitude toward today’s trends and technology. Perhaps provincial nightclubs will host ‘2020 nights’ where punters will dance ironically to Taylor Swift and Drake songs wearing Crocs and hoodies, supping on White Claws and making heart emojis with their fingers. I’m not sure what the product portfolio of a big tech company will be in 20 years, but I’m almost certain it will involve some ‘new things’. So what might they be, and what might they do

As We Look to the Future 

When we consider where we are today with our devices, we can look to Vannevar Bush's work on the Memex, which eventually influenced the development of the hypertext system and led to the creation of the World Wide Web. All human written knowledge from books, records, and communications is in our tiny devices. However, I believe the concept behind these devices is now outdated. Our devices are designed for use by a single person, focusing on a single interaction that lasts only a moment, 10 seconds. We currently have no context for these interactions. We don't have a clear sense of a tool that could help or build a community that spans all of us. What are the most relevant activities and interactions with my people, groups and societies that a tool could help with? What are the most relevant shared representations of reality that a new tool should be built on, and what value would that create for people, groups and societies? An ethnographic field guide would be helpful to understand how we could have a more profound impact on an individual's life and then impact their wider community. We would need to examine the social ecology base in the UK, with a representative mix across ages (18-25, 25-45, and 45+), gender, race, ethnicity, and occupation.

Human + Digital Companion > Great Human

We are missing a few core elements for this next development and to achieve it, we are going to make gradual progress. There will be systems that learn from videos about how the world works and then make good representations, but it will take time to match human scale and performance. There will be systems that have large associated memories so they can remember things, but not by tomorrow. There are a lot of new techniques to develop, and it will take time to get them to work together as a full system. 

Moravec's Paradox suggests that the skills humans take for granted, such as perception and mobility, result from incredibly complex biological processes that are difficult to replicate in artificial systems. Tasks we consider difficult, like advanced calculations, often have straightforward algorithms that computers can execute efficiently.

Recreating any human skill is a formidable task, considering the time and effort it takes to evolve in animals. Our oldest, largely unconscious and seemingly effortless skills have been honed over a significant period, making them particularly challenging to replicate in artificial systems. 

Examples of these skills that have taken millions of years to evolve are recognising a face, moving around in space, judging people’s motivations, and catching a ball. Recognising a voice, setting appropriate goals, and paying attention to exciting things all involve perception, attention, visualisation, motor skills, social skills, etc. More recently, examples have appeared in mathematics, engineering, games, logic, and scientific reasoning. These skills and techniques were acquired recently, in historical time, and have had at most a few thousand years to be refined, primarily by cultural evolution. 

There has always been optimism about the birth of artificial intelligence. Moravec’s Paradox results from realising that understanding reality or the world is not as easy as we think. “It is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility."

Intelligence is not linear and cannot be measured with a scale or a single number, as we attempt to do with IQ.  Are humans smarter than orangutans? In some ways, yes, but in other ways, orangutans are smarter than humans in many domains, which allows them to survive in the forest. 

IQ is a very limited measure of intelligence and is only relevant to humans. It only measures one type of ability that may be relevant for some tasks but not others and some beings but not others. The basic tasks an orangutan finds easy are different from those humans find easy. Using IQ to compare us as ‘intelligent entities’ has no relevance. Intelligence is a collection of skills and an ability to acquire new skills efficiently. The collection of skills that a particular intelligent entity possesses or is capable of learning quickly is different from the collection of skills of another one. You can’t simply measure and compare two things to decide whether one is more intelligent;  it’s multi-dimensional.

Curb Cut Effect

Innovation is exciting. It presents new possibilities and ideas that could start conversations and solve real problems people face today. Something I think is valuable more than ever is the curb-cut effect, a phenomenon whereby innovations designed for people with disabilities have benefited a wider range of people. Closed captioning is an example of this. Is it also being considered for technologies like smart glasses or voice assistance? Many of these new products could still be in the lab and we could consider them still in their infancy but we should take them seriously. Innovating openly and transparently will help push humanity forward. For the first generation, designers and even the end user are still working out where these products will fit, their purpose, and how they will prove valuable to people. We must foster an environment that allows mistakes to happen and has a feedback loop. Almost every business, person, and location in the world can contribute enormously if we are open, honest, and transparent about what problems we are trying to solve or the wild ideas we are trying to build. These products will transform our lives, and some will fail. Because of this, we need to rethink the language we use when communicating what these finished products will be. As I discuss with my friends, family, and peers where we are today and what we hope for the future, I urge us to be brave and bold but also clear about the grand challenges of today and the time scales to solve them. Bombastic claims, optimistic projections, and energetic positivity make me anxious. 

Are we communicating our innovations well?

Society's respect for some of the new technologies launched has pushed some toxic conversations. We need society to engage positively with exploration and innovation; we must change our thinking. While we can't simply build things that are not flawed, this should not stop us from building new things and dreaming of a bright future. Innovation will take time, but we have to be optimistic about the future and launch these new things with a more open and honest conversation. This might help us move forward from the negative sentiment that always draws us away from innovating.  If we rethink marketing and the model for communicating ‘new ideas’, this might help us change how the public thinks about technology. It could enable us to have a more productive culture around technology, be more innovative, try new things, and bring us all more joy.  

Hope for the Future, Human + AI > Great Human.

I’d like to advocate for the perception of computers as social actors and a framework where technology interacts with us personally, leveraging context and specialised knowledge beyond what's available through a simple internet search. This vision extends to devices like smart glasses or AI-powered cameras that possess a contextual understanding of the user and their environment, offering personalised assistance and recommendations. Imagine a device that knows your health metrics and the context of your activities, allowing it to suggest tailored health advice or adjustments to your routine.

We need to envision the next wave of technology, such as smartwatches, pins, or glasses, not merely as tools for optimisation but as integral components that enhance our daily lives through advanced sensory perception and contextual understanding. This approach aims to move beyond the simplistic tracking of objects to a more nuanced interaction with our environment and technology.

Why simple queries are not the future

Imagine asking a box on a pillar at Home Depot, “Where are the nails?” and getting directions, or your fridge responding with helpful advice when you say, “Why is the ice maker broken?” or your car answering, “How do I change the wiper speed?”. I think of these voice assistants for everyday objects as “A New Digital Companion”. Agents are great at answering questions, but only in a specific domain. I want them in my life, but they don’t yet exist. If they’re as helpful as I think, why aren’t they already here and why is now the right time for them to succeed? We need to move beyond these simple queries of our physical world. 

What is “A New Digital Companion”?

I’m a strong believer in computers as social actors and the idea that people want to interact with new technology as if it were another person. We ask people questions rather than googling on our phones because, unlike a search engine, the other person has a lot of context and specialised knowledge. With this in mind, I want to consider what would happen if we had a pair of smart glasses or a small AI-powered camera with a contextual understanding of us and the real world. What would this little assistant bring to our lives if it had a complete model of our physical world and daily lives? 

LLM, with a contextual understanding of our physical environments and daily lives, would have the capacity to make helpful recommendations beyond just being used as a replacement for Google search. If, for example, my Agent knows my heart rate, other health details, and the environment I’ve been running in, could it recommend a different type of trainer or ask me to shorten my running stride? 

Sensory anthropologists study how social structures determine the use of the senses and the meaning of the resulting perceptions (Hoes, 1991). In our scientific society, with its emphasis on physical explanations, the categories for sensing the external world are mostly sorted by the combination of biological organs and physical stimuli: ears are for hearing sound, eyes are for seeing light, and skin is for touching surfaces. Yet even with this bias toward concrete labelling, our culture does not notice the many different kinds of information processing that compose a single sensory modality. For example, the tactile modality-touch includes independent sensors for vibration, texture, temperature and movement. Aural architecture and auditory spatial awareness provide a way to explore our aural connection.

Conclusion

This gives a general view of what I'd like to do next and how we need to move beyond where my keys are to track aboard. It is a static fantasy or performance in daily life. So, the next iteration of smartwatches, pins, or glasses is helpful, not just for optimising us. We need to focus on the hard problems and be scientists with an engineer’s mindset. Spoken words limit us regarding what we need to prototype, make and build, so we need to be concrete rather than talk about the problem. We have no idea what professions will be hot 10 or 15 years from now. In the same way, who could have thought 20 years ago, or even five years ago, that the hottest job was mobile app developer? Smartphones hadn’t been invented.

So, in the words of Arthur C Clark, “The only way to discover the limits of the possible is to go beyond them, into the impossible.” 

Thanks for Reading.

— Q