Microsoft wants to build artificial general intelligence: an AI better than humans at everything

Kelsey Piper 
www.vox.com
Originally published July 22, 2019

Here is an excerpt:

Existing AI systems beat humans at lots of narrow tasks — chess, Go, Starcraft, image generation — and they’re catching up to humans at others, like translation and news reporting. But an artificial general intelligence would be one system with the capacity to surpass us at all of those things. Enthusiasts argue that it would enable centuries of technological advances to arrive, effectively, all at once — transforming medicine, food production, green technologies, and everything else in sight.

Others warn that, if poorly designed, it could be a catastrophe for humans in a few different ways. A sufficiently advanced AI could pursue a goal that we hadn’t intended — a recipe for catastrophe. It could turn out unexpectedly impossible to correct once running. Or it could be maliciously used by a small group of people to harm others. Or it could just make the rich richer and leave the rest of humanity even further in the dust.

Getting AGI right may be one of the most important challenges ahead for humanity. Microsoft’s billion dollar investment has the potential to push the frontiers forward for AI development, but to get AGI right, investors have to be willing to prioritize safety concerns that might slow commercial development.

The info is here.

Not all who ponder count costs: Arithmetic reflection predicts utilitarian tendencies, but logical reflection predicts both deontological and utilitarian tendencies

NickByrdPaulConway
Cognition
https://doi.org/10.1016/j.cognition.2019.06.007

Abstract

Conventional sacrificial moral dilemmas propose directly causing some harm to prevent greater harm. Theory suggests that accepting such actions (consistent with utilitarian philosophy) involves more reflective reasoning than rejecting such actions (consistent with deontological philosophy). However, past findings do not always replicate, confound different kinds of reflection, and employ conventional sacrificial dilemmas that treat utilitarian and deontological considerations as opposite. In two studies, we examined whether past findings would replicate when employing process dissociation to assess deontological and utilitarian inclinations independently. Findings suggested two categorically different impacts of reflection: measures of arithmetic reflection, such as the Cognitive Reflection Test, predicted only utilitarian, not deontological, response tendencies. However, measures of logical reflection, such as performance on logical syllogisms, positively predicted both utilitarian and deontological tendencies. These studies replicate some findings, clarify others, and reveal opportunity for additional nuance in dual process theorist’s claims about the link between reflection and dilemma judgments.

A copy of the paper is here.

First do no harm: the impossible oath

Kamran Abbasi

BMJ 2019; 366
doi: https://doi.org/10.1136/bmj.l4734

Here is the beginning:

Discussions about patient safety describe healthcare as an industry. If that’s the case then what is healthcare’s business? What does it manufacture? Health and wellbeing? Possibly. But we know for certain that healthcare manufactures harm. Look at the data from our new research paper on the prevalence, severity, and nature of preventable harm (doi:10.1136/bmj.l4185). Maria Panagioti and colleagues find that the prevalence of overall harm, preventable and non-preventable, is 12% across medical care settings. Around half of this is preventable.

These data make something of a mockery of our principal professional oath to first do no harm. Working in clinical practice, we do harm that we cannot prevent or avoid, such as by appropriately prescribing a drug that causes an adverse drug reaction. As our experience, evidence, and knowledge improve, what isn’t preventable today may well be preventable in the future.

The argument, then, isn’t over whether healthcare causes harm but about the exact estimates of harm and how much of it is preventable. The answer that Panagioti and colleagues deliver from their systematic review of the available evidence is the best we have at the moment, though it isn’t perfect. The definitions of preventable harm differ. Existing studies are heterogeneous and focused more on overall rather than preventable harm. The standard method is the retrospective case record review. The need, say the authors, is for better research in all fields and more research on preventable harms in primary care, psychiatry, and developing countries, and among children and older adults.

Veil-of-Ignorance Reasoning Favors the Greater Good

Karen Huang Joshua D. Greene Max Bazerman

PsyArXiv
Originally posted July 2, 2019

Abstract

The “veil of ignorance” is a moral reasoning device designed to promote impartial decision-making by denying decision-makers access to potentially biasing information about who will benefit most or least from the available options. Veil-of-ignorance reasoning was originally applied by philosophers and economists to foundational questions concerning the overall organization of society. Here we apply veil-of-ignorance reasoning in a more focused way to specific moral dilemmas, all of which involve a tension between the greater good and competing moral concerns. Across six experiments (N = 5,785), three pre-registered, we find that veil-of-ignorance reasoning favors the greater good. Participants first engaged in veil-of-ignorance reasoning about a specific dilemma, asking themselves what they would want if they did not know who among those affected they would be. Participants then responded to a more conventional version of the same dilemma with a moral judgment, a policy preference, or an economic choice. Participants who first engaged in veil-of-ignorance reasoning subsequently made more utilitarian choices in response to a classic philosophical dilemma, a medical dilemma, a real donation decision between a more vs. less effective charity, and a policy decision concerning the social dilemma of autonomous vehicles. These effects depend on the impartial thinking induced by veil-of-ignorance reasoning and cannot be explained by a simple anchoring account, probabilistic reasoning, or generic perspective-taking. These studies indicate that veil-of-ignorance reasoning may be a useful tool for decision-makers who wish to make more impartial and/or socially beneficial choices.

The research is here.

Dante, Trump and the moral cowardice of the G.O.P.

Charlie Sykes
www.americamagazine.com
Originally published July 21, 2019

One of John F. Kennedy’s favorite quotes was something he thought came from Dante: “The hottest places in Hell are reserved for those who in time of moral crisis preserve their neutrality.”

As it turns out, the quote is apocryphal. But what Dante did write was far better, and it came vividly to mind last week as Republicans failed to take a stand after President Trump’s racist tweets and chants of “Send her back,” directed at Representative Ilhan Omar of Minnesota, who immigrated here from Somalia, at a Trump rally in North Carolina.

In Dante’s Inferno, the moral cowards are not granted admission to Hell; they are consigned to the vestibule, where they are doomed to follow a rushing banner that is blown about by the wind.

(cut)

Despite some feeble attempts at rationalization, there was clarity to the president’s language and his larger intent. Mr. Trump was not merely using racist tropes; he was calling forth something dark and dangerous.

The president did not invent or create the racism, xenophobia and ugliness on display last week; they were all pre-existing conditions. But simply because something is latent does not mean it will metastasize into something malignant or fatal. Just because there is a hot glowing ember does not mean that it will explode into a raging conflagration.

The info is here.

Ethics and automation: What to do when workers are displaced

Tracy Mayor
MIT School of Management
Originally published July 8, 2019

As companies embrace automation and artificial intelligence, some jobs will be created or enhanced, but many more are likely to go away. What obligation do organizations have to displaced workers in such situations? Is there an ethical way for business leaders to usher their workforces through digital disruption?

Researchers wrestled with those questions recently at MIT Technology Review’s EmTech Next conference. Their conclusion: Company leaders need to better understand the negative repercussions of the technologies they adopt and commit to building systems that drive economic growth and social cohesion.

Pramod Khargonekar, vice chancellor for research at University of California, Irvine, and Meera Sampath, associate vice chancellor for research at the State University of New York, presented findings from their paper, “Socially Responsible Automation: A Framework for Shaping the Future.”

The research makes the case that “humans will and should remain critical and central to the workplace of the future, controlling, complementing and augmenting the strengths of technological solutions.” In this scenario, automation, artificial intelligence, and related technologies are tools that should be used to enrich human lives and livelihoods.

Aspirational, yes, but how do we get there?

The info is here.

Ethics working group to hash out what kind of company service is off limits

Chris Marquette
www.rollcall.com
Originally published July 22, 2019

A House Ethics Committee working group on Thursday will discuss proposed regulations to govern what kind of roles lawmakers may perform in companies, part of a push to head off the kind of ethical issues that led to the federal indictment of Rep. Chris Collins, who is accused of trading insider information while simultaneously serving as a company board member and public official.

(cut)

House Resolution 6 created a new clause in the Code of Official Conduct — set to take effect Jan. 1, 2020 — that prohibits members, delegates, resident commissioners, officers or employees in the House from serving as an officer or director of any public company.

The clause required the Ethics Committee to develop by Dec. 31 regulations addressing other prohibited service or positions that could lead to conflicts of interest.

The info is here.

Ethical considerations in assessment and behavioral treatment of obesity: Issues and practice implications for clinical health psychologists

Williamson, T. M., Rash, J. A., Campbell, T. S., & Mothersill, K. (2019).
Professional Psychology: Research and Practice. Advance online publication.
http://dx.doi.org/10.1037/pro0000249

Abstract

The obesity epidemic in the United States and Canada has been accompanied by an increased demand on behavioral health specialists to provide comprehensive behavior therapy for weight loss (BTWL) to individuals with obesity. Clinical health psychologists are optimally positioned to deliver BTWL because of their advanced competencies in multimodal assessment, training in evidence-based methods of behavior change, and proficiencies in interdisciplinary collaboration. Although published guidelines provide recommendations for optimal design and delivery of BTWL (e.g., behavior modification, cognitive restructuring, and mindfulness practice; group-based vs. individual therapy), guidelines on ethical issues that may arise during assessment and treatment remain conspicuously absent. This article reviews clinical practice guidelines, ethical codes (i.e., the Canadian Code of Ethics for Psychologists and the American Psychological Association Ethical Principles of Psychologists), and the extant literature to highlight obesity-specific ethical considerations for psychologists who provide assessment and BTWL in health care settings. Five key themes emerge from the literature: (a) informed consent (instilling realistic treatment expectations; reasonable alternatives to BTWL; privacy and confidentiality); (b) assessment (using a biopsychosocial approach; selecting psychological tests); (c) competence and scope of practice (self-assessment; collaborative care); (d) recognition of personal bias and discrimination (self-examination, diversity); and (e) maximizing treatment benefit while minimizing harm. Practical recommendations grounded in the American Psychological Association’s competency training model for clinical health psychologists are discussed to assist practitioners in addressing and mitigating ethical issues in practice.

First Steps Towards an Ethics of Robots and Artificial Intelligence

John Tasioulas
King's College London

Abstract

This article offers an overview of the main first-order ethical questions raised by robots and Artificial Intelligence (RAIs) under five broad rubrics: functionality, inherent significance, rights and responsibilities, side-effects, and threats. The first letter of each rubric taken together conveniently generates the acronym FIRST. Special attention is given to the rubrics of functionality and inherent significance given the centrality of the former and the tendency to neglect the latter in virtue of its somewhat nebulous and contested character. In addition to exploring some illustrative issues arising under each rubric, the article also emphasizes a number of more general themes. These include: the multiplicity of interacting levels on which ethical questions about RAIs arise, the need to recognize that RAIs potentially implicate the full gamut of human values (rather than exclusively or primarily some readily identifiable sub-set of ethical or legal principles), and the need for practically salient ethical reflection on RAIs to be informed by a realistic appreciation of their existing and foreseeable capacities.

From the section: Ethical Questions: Frames and Levels

Difficult questions arise as to how best to integrate these three modes of regulating RAIs, and there is a serious worry about the tendency of industry-based codes of ethics to upstage democratically enacted law in this domain, especially given the considerable political clout wielded by the small number of technology companies that are driving RAI-related developments. However, this very clout creates the ever-present danger that powerful corporations may be able to shape any resulting laws in ways favourable to their interests rather than the common good (Nemitz 2018, 7). Part of the difficulty here stems from the fact that three levels of ethical regulation inter-relate in complex ways. For example, it may be that there are strong moral reasons against adults creating or using a robot as a sexual partner (third level). But, out of respect for their individual autonomy, they should be legally free to do so (first level). However, there may also be good reasons to cultivate a social morality that generally frowns upon such activities (second level), so that the sale and public display of sex robots is legally constrained in various ways (through zoning laws, taxation, age and advertising restrictions, etc.) akin to the legal restrictions on cigarettes or gambling (first level, again). Given this complexity, there is no a priori assurance of a single best way of integrating the three levels of regulation, although there will nonetheless be an imperative to converge on some universal standards at the first and second levels where the matter being addressed demands a uniform solution across different national jurisdictional boundaries.

The paper is here.