Driverless car makers could face jail if AI causes harm

AI technologies which harm workers could lead to their creators being prosecuted, according to the British government.

analytics anywhere

Makers of driverless vehicles and other artificial intelligence systems could face jail and multi-million pound fines if their creations harm workers, according to the Department of Work and Pensions.

Responding to a written parliamentary question, government spokesperson Baroness Buscombe confirmed that existing health and safety law “applies to artificial intelligence and machine learning software”.This clarifies one aspect of the law around AI, a subject of considerable debate in academic, legal and governmental circles.

Under the Health and Safety Act of 1974, directors found guilty of “consent or connivance” or neglect can face up to two years in prison.

This provision of the Health and Safety Act is “hard to prosecute,” said Michael Appleby, a health and safety lawyer at Fisher Scoggins Waters, “because directors have to have their hands on the system.”

However, when AI systems are built by startups, it might be easier to establish a clear link between the director and the software product.

Companies can also be prosecuted under the Act, with fines relative to the firm’s turnover. If the company has a revenue greater than £50 million, the fines can be unlimited.

The Health and Safety Act has never been applied to a case of artificial intelligence and machine learning software, so these provisions will need to be tested in court.

Source: Sky.com

Advertisements

7 Robotic Process Automation Pitfalls & How to Avoid Them

AnalyticsAnywhere

In order to implement RPA “the smart way” and take the most advantage of it, you should be aware of the robotic process automation pitfalls from the very beginning.

We must all acknowledge and take seriously the fallibility of our endeavours, and, consequently, we should not allow ourselves to indulge in surreal expectations. The simple truth is that sometimes projects fail, for a very large number of reasons. According to IBM Systems Magazine, up to 25% of technological projects fail downright, while up to half of them require extensive revisions by the time they are set to go.

Bernard Marr writes in Forbes adds that more than half of the technological failures are in fact due to poor management, and only 3% are caused by technical problems. DIfficulties may also arise from not choosing the right processes to automate.Here are 8 questions to ask that should ease decision making in this regard.

So let us now delve a little deeper into potential robotic process automation pitfalls and corresponding means to avoid them in the course of implementing RPA.

Forethought is definitely needed for good results and successful RPA implementation. Of course you should first spell out what ‘successful’ means; but for now, let us tackle the question about what could go wrong during the implementation of your software robots. Here is a list of 7 aspects that ought to be considered and/or avoided if you want to stay safe from robotic process automation pitfalls.

1. Not choosing the right processes to automate in the beginning

This refers to picking the process that is most appropriate for an effective start of using automation in your business. By no means should you neglect a thorough, exhaustive and, of course, realistic evaluation of the tasks that may be passed on to robots. You do not want to start automating the wrong things, resulting in difficult to manage workflows.

A piece of advice courtesy of Cem Dilmegani, CEO at appliedAI, is that you should consider features like, for example, the process complexity and its business impact. Briefly put, you should perform a cost – benefit analysis of automating the candidate processes, based on what you consider to be your top goals.

2. Trying to implement robotic process automation on your own

You probably know this by now, otherwise you wouldn’t be reading this: RPA provides highly technical ways to carry out faster and more efficiently the dull jobs that would cause your employees unnecessary distress, boredom and fatigue. Precisely because of the high level of technicality, it is not at all advisable that you attempt to carry out the implementation process on your own.

Division of labour is with us for good reasons, so you must not forget to delegate the responsibility of implementation to the specialists who can best handle it. Tony Warren, executive vice president, head of strategy and solutions management at FIS, mentions things like “technical maintenance, operational monitoring and the appropriate change management procedures” among the RPA features that call for the right level of expertise, which specialist implementation navigators possess.

3. Not setting clear objectives for your automation strategy

This is a more general rule of thumb: it is vital that your business objectives, as well as the role that you expect RPA to play in getting there, are crystal clear.

What do you need RPA for?

Relatedly, which software provider is likely to do the best job for what you need?

While uncertainties in these respects are likely to be burdensome, definite answers to such questions will facilitate a smooth transition to delegating the tedious, repetitive tasks in your business to software robots.

4. Not having a “bird’s eye view” over the implementation process

As you probably know by now, RPA implementation is a complex enterprise. In fact, this comes as no surprise for an activity meant to take such deep effect on your business. So in order to achieve your goals, you need to ensure proper executive control.

This requires a group or an individual who can watch over the whole process from the top, so to say. Some call this essential aspect “operational oversight”, others – “governance of accretion” or simply “governance”, while others emphasize how important it is to include in the responsible team not only domain-specific specialists but also someone to take over the executive role of “central process unit”. In the long run, this can take the form of a robotic process automation centre of excellence that warrants a strategic maintenance of the system.

5. Not ensuring the scalability potential of your software robots

Scalability is one hidden gem that is certainly responsible for the larger-scale adoption of RPA. Which means that you really should not allow anything to stand in the way of scalable bots that can ensure consistent, across-the-board use of RPA in individual departments of your business.

6. Relying solely on the IT department

You certainly do not want to condition the smooth running of your automated processes to the IT department. Of course, it goes without saying that IT assistance is necessary for automation, but the idea is that you should not overdo it.

The bottom line is something along the lines of the phrase ‘render unto IT the things that are intrinsically IT-related (e.g. automation codes), and unto other departments the things that are better dealt with by other departments’. As Schultz puts it, “finance cannot depend on IT for RPA; it needs to be owned by the business side.”

7. Not testing your software robots thoroughly

Even if you may not like the phrase ‘haste makes waste’ after having heard it one million times, you have to admit there is some truth to it. And since you do not want to waste the effort, time, money and hope that you invested in RPA, you also do not want to stumble at the threshold.

As our own Daniel Pullen puts it, you need to test processes in production prior to full go-live to ensure there is a like-for-like behaviour between Dev and Production. This includes ensuring the applications are the same version, testing applications under normal and peak loads throughout the day, servers & applications in a server farm all behave identically (both operation and speed), etc.

Conclusion

We believe that you are now better prepared to embark on a successful RPA journey. Failure anticipation is not meant to alarm you, rather to motivate you to have a realistic view over what might happen so that you can prevent the pitfalls.

Anticipating and planning pro-actively should take you a step closer to gloriously passing the finish line. Although the word ‘finish’ is not perfectly fit here, since what you aim for with robotic process automation is a long-term sustainable development of your enterprise. As UiPath puts it, with “a comprehensive understanding of your company’s automation needs and the value proposition RPA provides, you can ensure a successful RPA implementation scheme that is both cost-effective and timely”.

Such extensive understanding can lead you to make use of the best practices for robotic process automation implementation. Wisely selecting the processes, a plain understanding of the required human resources or reliance on an ‘RPA sponsor’ are some of those practices, on which you can read more here.

Source: cigen.com.au