Data and its impacts on workers and citizens

Much of the discussion around new technology focuses around things like industrial and auxiliary robots, remotely-operated machinery, and perhaps increasingly, some types of autonomous vehicles.

But for many workers the main challenge will be adapting to working alongside and perhaps under the ‘management’ of machines and algorithms in the context of a more connected economy that runs on data. Data will transform power relations between employers and employees, between elites and the people, and between states too. In this sense the production and use of data marks a fundamental shift and a huge political challenge for workers.

To deal with this challenge unions need to push for the nationalisation of data and tech companies, the creation of publicly-owned tech infrastructure and the regulation of national digital spaces.

But what is data and how is it ‘made’?

Data is information.[1] Some of it is produced by sensors that measure the individual component operations of work processes. So in a delivery truck there are sensors that measure eye movements, hand movements, as well as braking and accelerating and the use of gears. There are also sensors that measure distance travelled and routes taken. In a warehouse there are sensors that measure hand movements, distance travelled, time taken to complete an operation, and number of packages dealt with. In a port crane there are sensors that measure wind speed, weight of container, speed of travel and so on.

Some of this data is produced by people. For example, we produce data through our digital or online activities (especially through our mobile phones and social media use) and we produce data at work while we labour ‘connecting’ the various parts of a work process. Algorithms can use a lot of this digital data to make comparisons. So for example, an algorithm can collect information on how long it takes a worker to answer an email or a phone call, and then use it to compare them to others doing a similar job. And we also produce data as we wander around with our mobile phones. This data can be used to measure flows of people across a transport network, or through a building. Similar data can be produced by tablets or other tools with embedded GPS chips or RFID tags. This allows workers to be tracked anywhere in a building or workplace.

So from the workers’ perspective data can be understood as information gathered on people and processes. This data is produced either by people themselves during their digital interactions (whether at work or not), or by monitoring equipment that ‘watches’ people and tools.

The resulting data can be used to produce a picture of the functioning of an individual, or of a workplace, or a society, from whichever aspect you wish to describe — work, leisure, education, health etc.

There are clear privacy implications here, as well as less obvious implications for power relations in the workplace and in society more broadly.

Data and power

Many experts argue that data will be a key ‘material’ in the digital economy, something to be traded and profited from. The widespread use of the internet, social media and of sensors and their increasing interconnectivity is already producing vast quantities of data, with more produced every day.

Data is the raw material of knowledge. As such it is a powerful tool. In a connected economy data can arguably become a form of capital, something that can be traded for profit.

Since technology usually acts to deepen existing trends in society, in an unequal society data will serve to further deepen inequality. It will also become another aspect of inequality.

People with power will decide what data gets collected about whom and what this information gets used for, whose problems it resolves. For example, data today is mostly being collected about workers, and not about business people or tax havens. That is a choice about data use that stems from existing inequality.

Therefore some people will have every aspect of their work and leisure (and potentially their health) measured and analysed by others, while some will have the power to prevent this without their permission. Some people will be able to use data, and others will not.

The issue with data is therefore who ‘owns’ it and who decides what is done with it. The various types of data being collected allow the data controllers to identify commonalities across groups of people, creating new ‘communities’, but also allowing groups to be ‘targeted’ — whether for advertising or for healthcare or perhaps for more sinister aims. One such group could be members of trade unions for example, or those likely to become trade unionists. So data is a political issue.

There are increasingly sophisticated tools available to analyse data. However, at the moment both the data and the tools that can analyse them are in private hands — largely the preserve of the ‘Big Five’ digital companies — Google, Apple, Facebook, Amazon and Microsoft (the GAFAM). And the decisions around what data can be used for are being made by them. And their criteria are to use data in their struggle against other data and tech companies, not to use it to resolve social or macroeconomic problems.

Much of the data being produced today is being used to create more effective forms of marketing and stimulation of consumption.

The collection, storage and sale of data is therefore becoming big business, and access to data is what drives the value of these five US companies in particular. But there are also many smaller tech companies being hired by employers to help them improve productivity. And much of this revolves around the collection and use of data in the workplace.

However, we have no idea what else data companies might use our data for. And we have no influence over it either, particularly once it is stored somewhere beyond our jurisdiction. The recent scandal over Facebook and Cambridge Analytica has highlighted the political importance of this issue.

Data in the workplace

In the workplace data is being used to resolve problems for employers. It could be used to make work processes more pleasant, or more stimulating, but in capitalism technology is usually used to make a process cheaper, not safer or more pleasant, since there is no money in that.

The data collected while people work describes the work process as a whole, but it also describes the person, or people, performing it. The data gathered can be used to identify ‘efficiencies’ in the process, or it can be used to improve how people work — but from the perspective of employers, improvement generally means doing more in less time. Data can also be used be manipulated to incentivise certain behaviours, as Uber software engineers do with drivers.

The problem for workers is that they generally have no participation in deciding what technology should be used in the workplace, nor any say in what it does. Nor do they own the data, despite it being ‘produced’ by them while they are working.

Currently workplace data, largely produced by employee monitoring, is mainly used to help discipline labour so as to help managers force workers to do more in less time. Software can monitor email response times, for example, or track sales or the location of a worker throughout the workday and so on. Algorithms can then benchmark workers against performance indicators, or against each other, or however it is decided.

In a sense, work now consists of two elements — the work process itself, and the data that the worker produces about this process, and about themselves as a worker. Workers clearly have a claim to at least shared ownership of this type of data because it would not exist without their efforts, even if it is a by-product. There is no legal framework around this at the moment, and no incentives for companies to use technology in ways that do not prioritise profit over the well-being of their workers.

As a result, data threatens to massively boost the already overwhelming power of employers in the workplace. For example, in some cases microphones and email monitoring are used against union organisers, and during disputes electronic passes can be used to create instant lists of strikers.

In addition, developments in Artificial Intelligence algorithms are potentially creating further challenges. AI uses large sets of data to answer questions. AI can now be used in order to analyse biometric data, facial expressions and voice tone in order to assess physical and mental well-being. In combination with other data it can be used to create personality profiles. These could be used during the hiring process. In the workplace these could be used to prevent organising, or to pre-empt disputes by sacking unhappy workers. In combination with social media data this type of information could be used by companies to identify union members, or even those workers likely to become union members, or assist in organising efforts. In other words, AI threatens to exacerbate existing trends and become a highly imperfect digital overseer of human labourers.

Across the world workers are already experiencing employer use of social media to undermine organising efforts, either by using false profiles to monitor Facebook groups or using it to harass and threaten workers involved in industrial action. Can we guarantee that social media platforms do not sell worker data to employers for this purpose? The Cambridge Analytica example would argue that we cannot.

At the moment neither the digital titans nor the smaller data companies are covered by any regulations on what they can and cannot use worker data for. Nor are there any ethical guidelines around the uses of AI in the workplace. These are discussions that workers need to be part of.

Data in society

But there is no use trying to control the collection, ownership and uses of data in the workplace if we don’t acknowledge that this is a broader social issue.

Just like it does in the workplace, data collection, ownership, analysis and use, threaten to increase the already immense power of wealthy elites. Across the world people are beginning to realise that their data is being collected and used without their having any control over this process. And they are starting to ask questions.

Data is knowledge and knowledge is power. Whoever collects, holds and uses the masses of data produced by people as they interact with the digital world becomes immensely powerful. Commercial data collected and used by companies also becomes hugely valuable.

Because of this, data becomes a political issue within countries, but also between them. Most of the world’s data is allegedly collected by digital titans from the United States. Most of the other data companies are also from the developed world. This raises the spectre of digital imperialism — what control do people across the world have over who collects their data and what their data is used for? And more worryingly, can they guarantee that it will not be used to disempower them? At the moment they cannot.

In the developed world the issue of data is usually seen from the perspective of privacy. What rights do people have over what their data is used for? This is the perspective that informed the French initiatives on data in the EU, and subsequently GDPR. But privacy is a social issue as well as one of individual preference. Data allows the identification of communities of interest, even before they begin to organise. Data is therefore highly political. The concerns around the collection and use of data in the EU were sparked by the Snowden revelations of US intelligence agency spying on European and Japanese political leaders and businesses. People asked how it could be right that US digital titans paid no taxes in the EU, but were free to collect vast amounts of data about EU citizens? Privacy concerns are exacerbated by the increase in various forms of terrorism which are used to justify ever increasing monitoring and surveillance.

Data is not just about what we produce as we browse the internet, or as we walk around with our mobile phones pinging. Data, in the form of information, is also something we consume. While we do this, we produce information about what we read, what we watch and how we feel about the issues of the day. This is some of the information about us that is collected by the digital titans. They watch everything we do in the digital world, and they use it to shape what we want, what we think is important. So the digital titans and social media also play a huge role, alongside traditional media, in shaping the information available to the public, and thereby shaping their sense of what is important and what isn’t.

They therefore have immense potential to control the issues that set the political and social agenda. We can see this in action today, in the debates around ‘fake news’ and how people in much of the world are scandalised by events in Syria, but largely agnostic on events in Yemen. We are encouraged to vilify Venezuela, but ignore systemic violence in Colombia. Access to billions of people across the world means that digital titans can now influence political agendas across the world very rapidly by creating transnational interest groups. In this sense, these companies should not just be considered digital, or even data companies. They are privately-owned political agenda setters.

This makes data a weapon too. As William Perry the US Secretary of Defence noted in the 1990s, ‘We live in an age that is driven by information. Technological breakthroughs . . . are changing the face of war and how we prepare for war.’[2] Why? Because the more you know about your opponent the easier it can be to identify weak points and then exploit them.

In the workplace data can be weaponised by the employer against the worker. In society it can be weaponised by elites against organised challenges to their control. Companies can also use steal each other’s data in order to outcompete one another — it is a form of industrial espionage — a practice as old as capitalism itself.

The only way to avoid abuses is to allow people, as workers and as a society, some form of control over their data.

Data in international relations

In the same way that data can be weaponised in the workplace or in society, it can be used in conflicts between countries too. In international relations data is becoming a new form of inequality. Some countries only produce data, other countries have the tech companies and people capable of analysing and using data. In this sense data reproduces the traditional inequality of capitalism between the centre and the periphery. This data inequality will exacerbate other forms of inequality and power imbalances within and between countries. It is also an issue of sovereignty, because of the way that data can inform the use of power. In international relations therefore, data can potentially be weaponised by those countries that can collect it, analyse it, and then use it.

And this is already happening. The use of social meta-data and information to effect political change is part of what is known as ‘information war’. Information war describes the way in which agenda-shaping can be used to force political change. In the words of the UK’s Joint Threat Research Intelligence Group it is ‘…using online techniques to make something happen in the real or cyber world.’ [3]

However, in order to use information war tools, it is necessary to understand the target society — what the strengths and weaknesses of that society are, and what people are angry about. You need to know what unites or divides people, how many people feel or think a particular way. In other words, you need data about that society. This data can be collected via social media platforms, which US intelligence officials have said is ‘revolutionising open source intelligence gathering’, but also through the export of digital governance systems which use biometric or DNA data and so on.[4] This data can then be used to underpin the other elements of information war — cyberwarfare, and psyops for example.

Today it is almost routine for countries to accuse each other of cyber-attacks (hacking), of spreading ‘fake news’, or supporting opposition abroad and of other forms of information war. The very vagueness of the term and the clandestine way these tools are used blurs the boundary between conflict and non-conflict. The distinction becomes clearer in areas where data has clearly been used to help organise protest.

In all countries the desire to control information flows has always meant a close relationship between media and intelligence agencies. This has not changed today. In the US for example, there are well known connections between leading tech companies, social media platforms and US intelligence agencies.[5] In 2011 for example, Google was involved in designing software that was intended to stimulate defections from the Syrian government.[6] There are no doubt similar connections between tech companies and intelligence agencies elsewhere too.

The Obama administration made the pursuit of ‘internet freedom’ abroad a goal of foreign policy, to support civil society organising abroad by allowing people free access to information, the right to publish their own media, and the right to freely communicate.[7] In light of US tech and social media companies’ dominance of the internet, this was clearly intended to strengthen US government ability to influence society in selected countries.

In the same year USAID set up a ‘Cuban-twitter’ called Zunzuneo, intended to gather private data for political purposes, and eventually to spread subversive messages in the hope of provoking a ‘Cuban Spring’.[8] Subsequently, social media has been used in this way to help organise and internationalise protest during the Arab Spring, and during Ukraine’s 2013–4 crisis, Venezuela’s ‘Guarimba’ protests, and also during recent violent protests in Nicaragua. Social data has therefore already been used in combination with other forms of information war in order to undermine and in some cases, bring down targeted governments.

The documented links between US intelligence and leading US tech companies highlights the potential for these social media to be used in order to pursue the goals of US foreign or commercial policy.[9] In contrast to the cases outlined above, it is notable how social media have sometimes NOT been used to internationalise protest — in Colombia, Honduras, Mexico, or Saudi Arabia for example.

In all these cases there were mass protests that were violently repressed by state forces, but the civil society twitter hashtags did not go viral in the anglosphere, and no international scandal of any note followed. It is unlikely that the US is alone in seeking to use social media and social meta-data in this way. NATO countries routinely accuse Russia of using ‘information warfare’, and many countries, including Britain, have set up cyberwar units dedicated to digital conflict and ‘information warfare’. The only difference is one of scale and access, with US social media companies having global reach and claiming to collect 70% of the world’s data. So the trend of militarising digital space will doubtless continue. In a world entering a period of international instability, this means that national data is not just a national raw material, it is a strategic national security issue.

These trends have led to some countries setting up controls on social media and online communications, and to prevent unlimited access to their national data by foreign tech companies. China, Cuba, Iran, Russia, Vietnam and others have set up firewalls and limited access to foreign social media. They oblige companies to hold data on their national territory. They argue that this provides them with ‘digital sovereignty’ against external intervention. Opponents argue that it enables social control. In a similar way, Ukraine has blocked access to Russian-language social media and internet browsers, and there are calls across NATO and Eastern Europe for controls over ‘Russian propaganda’. There is, therefore, a trend towards the fragmentation of the global internet into national, or at least regional spaces policed by military and intelligence agencies dedicated to preventing information being used against the status quo.

Left-wing governments have historically been subjected to economic, financial and other forms of pressure — whether we are talking about Wilson’s Labour government in the UK, the first Mitterand government in France, the Syriza government in Greece, the Chavez/Maduro government in Venezuela, or Allende’s Chile. There is therefore clear potential for information war techniques to be used against union-friendly governments elected with a mandate to limit corporate power and create a more equal society.

This makes data, and the unregulated collection of social meta-data a political issue for trade unions and their political allies.

In the same way that digital controls can help protect national interests in relation to external aggression, digital controls can also help protect commercial secrets or workers’ interests.[10] And we would argue that both commercial and workers’ interests are the essence of what is a truly ‘national interest’. If we want to protect commercial data, or workers’ rights against surveillance and the use of AI in the workplace, we need to protect and regulate our national digital spaces.

It is indicative that the United States is pushing to preserve a globalised digital space. It placed data and e-commerce at the heart of draft Trade in Services Agreement (TiSA) and at the December 2017 WTO meeting in Argentina.[11] The Trump administration has pushed for ‘a borderless, digitized global economy in which major technology, financial, logistics, and other corporations like Amazon, FedEx, Visa and Google can move labor, capital, inputs, and data seamlessly across time and space without restriction. They also want to force the opening [of] new markets, while limiting obligations on corporations to ensure that workers, communities, or countries benefit from their activities.’[12]

If countries sign up to the TiSA or if the WTO were to approve such measures it would lock in the unaccountable access of (mainly US) digital companies to data from across the world, with potentially serious consequences for workers because it would effectively prevent the development of national labour controls in future, and because it would subordinate national governments to mainly US transnationals. In other words, TiSA is the potential application of digital imperialism to both the developing and the developed world, subordinating workers even further to the interests of a handful of TNCs.

A globalised digital world that only benefits transnational corporations is not in the interests of workers anywhere, because it enshrines corporate privilege, does not protect workers from the downward pressure on wages created by offshoring and the internationalising of labour markets by digital platforms. Nor can it protect them from the effects of increasing surveillance, work monitoring and ‘Digital Taylorism’. The use of data by existing power structures can lock us in to the injustices of the current system.

Trade Unions need to develop a stronger understanding of the role of data/information in the development of political and social trends if they are to successfully defend workers’ rights, and if they are to lead the positive transformation of society during the 21st century.

The massive digital companies have been able to create and maintain an ‘ethical’ veneer (evident in them being called ‘digital unicorns’ — what could be less threatening than a unicorn?), yet many have begun to point towards the underlying similarity of their activities with those of the giant industrial corporations of the 20th century.

If data is a strategic raw material, and a key issue for national defence, then it should be treated in a similar way to other strategic resources. One way to at least limit the power of the digital titans is to enforce digital sovereignty so as to protect individual, group and collective data rights and their labour rights at national, or regional level. Companies that do not abide by national laws or pay national taxes should be blocked from the national digital space.

Another important step in neutralising the many threats to society posed by private ownership and control of data is to nationalise some big tech companies, to regulate the uses of data, or to create publicly-owned or controlled data or social media companies answerable to the public. Cloud services should be publicly owned to guarantee data security and the ethical use of information. We should also push for the creation of ‘national internets’ that protect key functions of the economy by working on different hardware than that used across the world today. Otherwise we cannot guarantee the secure functioning of a digital economy.

Countries also need to develop their high-tech sectors, particularly through publicly-owned tech companies, so as to avoid dependence on foreign technology, and provide people with the necessary skills to use technology to deal with the really big issues societies face today — climate change and inequality.

***

[1] There are several definitions of data. Data is information organised for analysis to help make decisions. In computer science it is numerical information suitable for computer processing. It is also information produced by sensors or organs that needs to be processed before use.

[2] See ‘Strategic Information Warfare’, RAND Corporation https://www.rand.org/pubs/monograph_reports/MR661/index2.html

[3] Information war is still loosely defined, but it is generally understood to combine electronic warfare, cyberwarfare and psychological operations in order to attack and exploit vulnerabilities in communications, spread disinformation and fear, and destabilise a target society. See ‘The “Cuban Twitter” Scam is a Drop in The Internet Propaganda Bucket’ and ‘How Covert Agents Infiltrate the Internet to manipulate, deceive and Destroy Reputations’, The Intercept, February 2014, https://theintercept.com/2014/02/24/jtrig-manipulation/

[4] See ‘US Military Plugs into Social Media for Intelligence Gathering’, The Wall Street Journal, August 2014, https://www.wsj.com/articles/u-s-military-plugs-into-social-media-for-intelligence-gathering-1407346557

[5] See ‘NSA has backdoor access to Internet companies’ databases’, CNET, June 2013 https://www.cnet.com/news/nsa-has-backdoor-access-to-internet-companies-databases/ or ‘NSA Prism programme taps into user data of Apple, Google and others’ The Guardian, June 2013 https://www.theguardian.com/world/2013/jun/06/us-tech-giants-nsa-data

[6] See ‘Google Planned to Help Syrian Rebels Bring Down Assad Regime’, The Independent, March 2016, https://www.independent.co.uk/life-style/gadgets-and-tech/news/google-syria-rebels-defection-hillary-clinton-emails-wikileaks-a6946121.html

[7] ‘The Political Power of Social Media’, Foreign Affairs, Jan-Feb 2011, https://www.foreignaffairs.com/articles/2010-12-20/political-power-social-media

[8] ‘US secretly created ‘Cuban Twitter’ to stir unrest and undermine government’, The Guardian, 14 April 2014, https://www.theguardian.com/world/2014/apr/03/us-cuban-twitter-zunzuneo-stir-unrest

[9] In 2001 the European Parliament published a report that investigated US and UK use of the Echelon communications monitoring system to carry out industrial espionage in the EU. In 2014 Edward Snowden also alleged that US intelligence services carried out industrial espionage against European companies.

[10] By digital controls we mean national or regional regulations aimed at controlling the use of, or access to, the Internet; the collection and use of data, and the uses of surveillance technology by either individuals, companies or government agencies.

[11] See Jane Kelsey, ‘TiSA: Not our future!’, Report for the IUF, February 2018.

[12] Deborah James, ‘State of Play in the WTO’, Counterpunch, June 2017.