"Just because you can do something doesn't mean you should," she says.

A 2016 study conducted by Avanade found that "78% believe that their organisation has not given enough thought to the workplace ethical dilemmas created by the increased use of smart technologies. Less than half of those interviewed (43%) have established guidelines on work impacted by the use of smart technologies or developed new roles in this area."

Adam-Gedge quotes Gartner's definition of digital ethics: "The system of values and principles a company may embrace when conducting digital interactions between businesses, people and things. Digital ethics sits at the nexus of what is legally required; what can be made possible by digital technology; and what is morally desirable."

Digital transformation means there is more access to data than ever before, providing additional opportunities to gain insights and to act on them – perhaps in an automated way.

One symptom of a lack of digital ethics is the way large volumes of data are ending up in the wrong places. This could be caused by simple human error, but there could be systematic issues.

Organisations should adopt a structured approach to projects, Adam-Gedge suggests, using multidisciplinary teams to consider the issues (including possible unintended consequences), rather than leaving them to software engineers and developers.

For example, driverless cars are getting a lot of attention recently, including the question of whether they would (or should) prioritise the safety of their passengers ahead of other people – what's the right choice between hitting a power pole and possibly killing the occupant, and hitting and possibly killing a pedestrian but allowing the occupant to walk away?

Another issue is people's expectation that data about themselves will be stored securely and privately, but this isn't always the case.

You might go along with allowing your car insurer to access data collected about the way you drive (eg, how rapidly you accelerate and brake, how quickly you take corners, what time of day you use the vehicle) in return for lower premiums if you're a relatively safe driver. If you are lead-footed, the insurance company might provide initially gentle feedback ("if you accelerated more gently you could reduce your fuel expenditure by $X a week") and if that doesn't work it might follow up with the stick: "Slow down or your premium will increase by $Y next year."

But what if the company's terms and conditions said it could share this data with its life insurance division or other insurers?

There are other aspects to doing the right thing. Consider services that perform DNA analysis in order to provide their customers with information about any health issues to which they may be predisposed. Even if a customer asks for the report to be sent by email, it may not be right to do that if it contains information that is better interpreted by a doctor, for example that they have a significantly higher risk of certain cancers.

Automation and related technologies can help customers get better service at the same time as saving money for the organisation, but it is important to consider the ethical side: "more needs to be done in that area," she says.

Insufficient attention to digital ethics can have a real downside for an organisation. "Brand damage is really significant."

Organisations report on, and talk about, ethics much more than they used to, not just as a compliance issue, but flowing from a realisation that employees, customers, investors and other stakeholders see ethical organisations in a more favourable light.

If an organisation is unable to attract and retain good talent, the relationship with its customers or clients will slip, and that leads to reduced financial performance.

So what should organisations do?

Adam-Gedge suggests starting at the top: work with the board and then work down. But she warns that it may be necessary to develop the board's capability for dealing with the technical aspects of ethics.

Then one should establish a digital ethics framework with multidisciplinary input. Just realise that "you can't think of everything or you'll never move."

That framework can then be applied to policy, guidelines and practices.

Digital ethics needs to be pervasive, not the preserve of the ethics team.

Culture will have an impact on digital ethics. For example, the culture in a heathcare organisation will likely be different to that in a manufacturing company. National cultures also vary in relevant ways. There is no single view of what is ethical, and different organisations have different risk appetites.

Organisations generally realise they need new and different skills in the ethics area, but the importance of skills development and acquisition should not be overlooked.

At the project level, she advocates a design-led approach — how to meet the needs of the particular persona — rather than focusing on technology specifics. "Most of the work we do is design-led rather than technology-led." So do a proof of concept, then a pilot, and so on. In line with contemporary thinking, it's better to fail fast rather than embark on a two-year project that turns out to miss an important consideration.

Automation can lead to opacity, so it is important to put ethical consideration at the forefront. But it isn't something that can be put aside once a project is rolling: you should keep asking about the about intended and unintended consequences for stakeholders.

If you're wondering whether it's worth all this effort, Adam-Gedge says the benefits of digital ethics are so great that organisations cannot afford to ignore them.