Big Data and The Great A.I. Awakening. What are the ethical and societal implications?
By Roberto V. Zicari, Frankfurt Big Data Lab, Goethe University Frankfurt
I drafted a note (almost a fragment) derived from various interviews I conducted on the subject of ethical and societal implications of AI and Big Data. This is work in progress, there is no claim to be perfect here nor precise. The goal is to increase awareness.
1. Data
It is becoming evident that data is an economic asset. Companies with big data pools can have great economic power [2].
Today, that shortlist would include Google, Microsoft, Facebook, Amazon, Apple and Baidu.
As an aside note, none of them are European…
But what is the interplay and implications of big data and artificial intelligence?
The (Big) data revolution has made the recent AI advances possible.
In this competing world of Artificial Intelligence (AI), what is more important, vast data pools, sophisticated algorithms or deep pockets?
One point of view is expressed by Andrew Ng, a Stanford professor who worked at GoogleX, co-founder of Coursera, and previously chief scientist at Baidu:
“No one can replicate your data. It’s the defensible barrier, not algorithms.”
“AI is akin to building a rocket ship. You need a huge engine and a lot of fuel. The rocket engine is the learning algorithms but the fuel is the huge amounts of data we can feed to these algorithms.”
We are in a new phase, technology is moving beyond increasing the odds of making a sale, to being used in higher-stakes decisions like medical diagnosis, loan approvals, hiring and crime prevention [2].
2. Societal implications
The consequence of this has been stated clearly by Steve Lohr [2]:
“The new, higher-stakes decisions that data science and AI tools are increasingly being used to make — or assist in making — are fundamentally different than marketing and advertising“.
“In marketing and advertising, a decision that is better on average is plenty good enough. You’ve increased sales and made more money. One stage in the life cycle of an emerging science. Marketing is a low-risk – and, yes, lucrative.”
The difference between using data for marketing and for other more critical decisions is explained in a crude way by Claudia Perlich, a top star German data scientists, currently working at an ad-targeting start-up in New York, Dstillery:
“What happens if my algorithm is wrong? Someone sees the wrong ad. What’s the harm? It’s not a false positive for breast cancer.”
These other decisions are practically and ethically very different. These are crucial decisions about individual people’s lives. Better on average isn’t good enough.
For these kinds of decisions, issues of transparency, accuracy, fairness and discrimination come into play.
3. Transparency
Let´s look at transparency. Transparency brings two issues with. First, do we want transparency? (for that, please see section 8. Motivation and conscience). Second, are we technically capable to open the black box of AI?
If there is no transparency, then trust in the AI system will become a leap of faith, which may not work that well.
Quoting Manuela Veloso, professor of computer science at Carnegie Mellon University:
“We need to be able to question why programs are doing what they do. If we don’t worry about the explanation, we won’t be able to trust the systems.”
We can argue for example that some sort of auditing tools are needed, and that the technology has to be able to explain itself, to explain how a data-driven algorithm came to the decision or recommendation that it did [2].
For example, a deep learning algorithm can be trained to perform a task, but it often operates in a black box, so we don’t know how decisions are made.
Not even data scientists really know how the most advanced algorithms do what they do. That could be a problem.[8]
4. Human in the loop
Machine autonomy raises specific ethical and safety concerns and regulation is a possible response.
In principle, we can use AI technologies either to automate or to augment humans.
An interesting point here is the wish for having a “human in the loop” for most of these kinds of decisions for the foreseeable future.
Ben Shneiderman a University of Maryland computer scientist and user interface designer has written eloquently on this point. His point is that it is essential to keep a human in the loop. If not we run the risk of abdicating ethical responsibility for system design.
One can argue that Ben Shneiderman suggestion: “it is essential to keep a human in the loop” – may be desirable, but in praxis may be unrealistic [7] .
For example, Google first version of self-driving car had a steering wheel and brake but in the latest version there is no steering or brake, only a big START/STOP button, because Google found that humans are not able to be always in the loop [7].
The key question here is: If something can be partially automated, and no regulations is in place, will it eventually be fully automated?
Which leads to the this other question: Are computer system designers the ones who will decide what the impact of these technologies are and whether to replace or augment humans in society?
Prominent technology leaders such as Tesla’s chief executive, Elon Musk, suggested we might need to regulate the development of artificial intelligence.
How difficult is it to reconcile the different interests of the people who are involved in a direct or indirect way in developing and deploying new technology?
For the designers of Intelligent Systems, how difficult is to draw a line between what is human and what is machine?
“This is why we have governments and governmental regulation. AI, in that respect is no different than any other technology. It should and can be regulated when human safety is at stake” [3].
The challenge here is to use these systems to enhance human thought, not for social control.
5. Ethical responsibilities
Viktor Mayer-Schönberger, Professor of Internet Governance and Regulation at Oxford University (UK), believes that when for instance we ask about the ethics of Big Data, it is wrong to focus on the ethics of algorithms, and much more appropriate to focus on the ethics of data use [5].
So, what are the ethical responsibilities of designers of intelligent systems?
Intelligent system designers do have ethical responsibilities. Quoting Oren Etzioni, Chief Executive Officer of the Allen Institute for Artificial Intelligence [4]:
“We have a profound ethical responsibility to design systems that have a positive impact on society, obey the law, and adhere to our highest ethical standards.”
6. Responsibility
If we delegate decisions to machines, who will be responsible for the consequences?
Oren Etzioni has no doubt:
“Of course we are responsible. That is already true today when we use a car, when we fire a weapon—nothing will change in terms of responsibility. “My robot did it” is not an excuse for anything ” [4].
7. Motivation and conscience
But I believe even a more important point in this discussion, is raised by Oren Etzioni when he talks about human motivations [4]:
“It is absolutely essential that we control the machines, and every indication is that we will be able to do so in the foreseeable future. I do worry about human motivations too. Someone said: I‘m not worried about robots deciding to kill people, I’m worried about politicians deciding robots should kill people.”
So, it boils down to this : The thing that motivates my actions will determine the direction I am going.
As stated in [1]: The individual (and collective) conscience is the existential place where the most significant things happen. Research, Change, Decision and Choice can take two diametrically opposite directions: can be either “pro or contra” the human person”
This means that we can use Big Data and AI to help people or if the intention is different, harm people.
So, data, AI and Intelligent systems are becoming sophisticated tools in the hands of a variety of stakeholders, including political leaders and others….
8. Risks
So what are concrete steps that can be taken to minimize and mitigate big data’s risk?
Viktor Mayer-Schönberger comments on this [5]:
“There is potentially too much at stake to delegate the issue of control to individuals who are neither aware nor knowledgable enough about how their data is being used to raise alarm bells and sue data processors.
There are many areas of modern life that are so difficult and intransparent for individuals to control that we have delegated control to competent government agencies.
We need decision makers and especially policy makers to better understand the power of Big Data.”
The goal is to ensure transparency, guarantee human freewill, and strike a better balance on privacy and the use of personal information.
9. Ethical use of data
Good data reflects reality and thus can help us gain insights into how the world works. That does not make such discovery ethical, even though the discover is correct [5].
Good intentions point towards an ethical use of data, which helps protect us again unethical data uses, but does not prevent false big data analysis. This is a long way of saying we need both, albeit for different reasons [5].
I started an initiative Data for Humanity [6], together with Andrej Zwitter (Professor at the University of Groningen) at the end of 2015, with the goal to: Bring people and institutions together who share the motivation to use data for the common good.
Data for Humanity calls for the use of five ethical principles for the use of data. The Five ethical principles when using data are:
Do no harm
Use data to help create peaceful coexistence
Use data to help vulnerable people and people in need
Use data to preserve and improve natural environment
Use data to help create a fair world without discrimination
I use Viktor Mayer-Schönbergerà quote as a concluding remark [5]:
” Big Data is at its core about understanding the world better than we do today.
We need more researchers team up with decision makers in charities, and more broadly civil society organizations (and the government) to utilize Big Data to improve our understanding of the key challenges that our society is facing.
Sources:
[1] “The good society and the future of jobs: Can solidarity and fraternity be part of business decisions?” MAY 8 -10, 2014 – Vatican City
[3] Machines of Loving Grace. Interview with John Markoff. ODBMS Industry Watch, August 11, 2016
[4] On Artificial Intelligence and Society. Interview with Oren Etzioni, ODBMS Industry Watch, January 15, 2016
[5] On Big Data and Society. Interview with Viktor Mayer-Schönberger, ODBMS Industry Watch January 8, 2016
[6] Data for Humanity: An Open Letter
[7] Gregory Piatesky`s comment on [3]
[8] The Dark Secret at the Heart of AI. MIT Technology Review, April 11, 2017