Last week I did some scene-setting about the internet and the environment, which will be a major theme at this year’s Internet Governance Forum (IGF). I chaired a session on this at the UK IGF this month.
This is the first time the environment’s been given prominence in global internet fora like the IGF, but it’s more than time that happened.
Digital technology has potential to improve energy efficiency (and other processes as well), which could contribute to a lower carbon future, but it’s also the fastest growing source of energy consumption (and so carbon emissions in the world today) – as well as one of the fastest growing sources of pollution.
A quick recap
Last week, I argued that the digital community needs to take that juxtaposition of desired and undesirable a lot more seriously – not to trade off risks and opportunities but to do three things:
maximise the potential value of digitalisation for environmental gain;
mitigate and minimise the environmental costs associated with it;
and monitor what’s happening so that we can do those other Ms effectively.
So what does that entail|? At the UK IGF, I suggested that four things were crucial.
Understanding what is happening
First, understanding’s crucial. As in other aspects of the digital society – employment, governance, rights, for instance – the discussion here’s inadequate. Digital optimists wax lyrical, with support from management consultants. Pessimists see doom ahead.
The fact is, we don’t know enough about what’s happening, and so we don’t think forwards with enough sophistication. We need more evidence, more analysis and foresight, less hype and spin, about the four kinds of impact that I mentioned last week:
direct effects resulting from the manufacturing of equipment, powering data centres and so on;
indirect effects arising from the use of new technologies, including (nowadays) the growing use of Internet of Things, AI and automated, internet-enabled, decision-making processes;
societal effects resulting from the changes we’re all making in the lives we live;
and rebound effects, which can see energy consumption rise as efficiency and prices fall.
Dialogue is also crucial, above all between digital insiders, policymakers and environmental experts.
Too many discussions about environmental impacts in fora like the IGF are dominated by internet insiders who think they know about environmental issues and are keen to offer what they call ‘solutions’. They should – as we did at this year’s IGF – start by asking environmental experts what they see as the challenges we face and then exploring what the digital sector is doing that is harmful as well as what it can contribute towards addressing problems.
But the same is true of global discourse on environmental issues. The impact of the digital society – for good or ill – has barely figured in global fora concerned with climate change.
That’s problematic. Climate change and digitalisation are two of the main factors – I’d add genetics and geopolitics – that look set to change our ways of living more than any others in the next two decades. The interplay between them’s fundamental; they need to be thought about together, not apart.
Third, we need the will to rethink digital governance.
The digital sector’s now the most powerful economic sector. Power in it’s concentrated in a few corporations with global reach and impact.
Decisions those corporations make today will define our future in ways over which we have no serious control. They’re located overwhelmingly in two countries – the United States and China – and they’re much less accountable than those in other powerful industries like oil and pharmaceuticals (which also have their problems).
The new technologies they work with are developing so fast that the impact of decisions that they take can become irreversible before we’ve worked out governance frameworks to deal with them, let alone whether they’re where we want to go in the first place.
Partly that’s because those corporations have grown up so fast; partly because our ideas of how to govern them have focused on potential rather than on risk; partly because control of them is in the hands of very few. We’ve given those few a lot of benefit of doubt, far more than we've allowed technologists and businesses in other sectors. How much better off we’d be today. I wonder, if we’d paid more attention in the early days of the internet to issues like cybersecurity and data privacy, rather than having to play catch up now.
Like its impact on rights, the internet's impact on the environment is too important to the public interest, now and in the future, to leave it to technologists and businesses. This matters just as much if the net environmental impact’s positive as if it’s negative.
The value of getting things right on the environment – in the public interest and that of future generations – is considerable, but so’s the risk of going wrong. That’s why my fourth proposal's for more responsible innovation. What might that mean in practice? I’ll give four examples.
First, the design of data centres. We should applaud the efforts that data corporations are making to derive data centre energy from renewables, but that’s not sufficient in the long term.
Using renewables in a growing number of expanding data centres means that renewables are less available for other users. Even if all data centres used renewables, as most may in ten years’ time, data volumes will still grow.
And they’re only part of the data value chain. Where data are generated and how they are transmitted are also involved. Improved efficiency in data gathering, analysis and management is crucial, as well as direct impacts running kit.
Designing technology and services
Second, the design of technology and services. Environmental considerations – especially energy requirements – should be integral to the development of standards for new technologies (like 5G and its successors) and in the design, development and deployment of services and applications.
We’ve an obvious example of the environmental (and commercial) damage that results when digital developers ignore this in the case of bitcoin: an innovation so profligate with energy that it already uses more than Switzerland – which means it’s neither scalable nor environmentally sustainable. The teeniest environmental audit would have spotted this beforehand.
Including environmental considerations in product and service design should be the norm rather than an afterthought.
Third, likewise, device design.
AI and IoT will lead to huge growth in the number of devices we all use. Churn – the turnover of devices – is crucial here. We change our digital devices more often than we change the other goods that we rely on. A new car might last us ten years (five if we are wealthy), and will then be bought and used by someone else. Phones are likely to be junked in three (two if we're geeky).
Short device lifecycles mean more energy consumption and more waste. Including environmental considerations in design, development and deployment – aiming for fewer devices which are more energy efficient, with more standardisation and collaboration between suppliers and longer product lifecycles – would mitigate environmental costs.
And fourth, e-waste. The scale of this is already beyond the capabilities of many governments to handle. Volumes are growing and will grow faster as new types of kit proliferate. Some digital waste is toxic and just 20% (at most) is yet recycled.
Continuing to dump dead kit in developing countries should be no-one’s answer. Industry, governments and international organisations need to work together much more effectively to address this before the digital detritus of today becomes the plastic of the future.
Those are four suggestions as to what’s required. Last week I called addressing these ‘greening the digital society’, but perhaps that’s not quite right; at its heart’s the goal of mitigating disbenefits associated with the benefits we hope to gain. Not greening then, more hoping that we get to stay on the right side of sustainability. Doing that’s less glamourous than launching a new iPhone every second year, but it’s also more important.
It adds, I think, to growing calls for the digital sector to enter into dialogue that’s not just multi-stakeholder but multi-faceted; to listen to expertise in areas beyond the digital before proferring ‘solutions’ to them; to recognise that with great power comes great responsibility, and that it should look to the future as well as maximizing bottom lines.
Next week: six challenges for internet governance in a world that’s digitising fast.