IEEE Technology and Society Magazine - June 2021 - 83

economic or financial logic that places profit-seeking
and the mitigation of legal risks above pro-social
concerns. When these motives shape the institutional
context, firms may simply lack the infrastructure, processes,
and culture to responsibly weigh social and
ethical concerns against economic ones. For example,
Hagendorff [14] argues that when companies are
driven by an economic logic, " engineers and developers
are neither systematically educated about ethical
issues, nor are they empowered, for example by
organizational structures, to raise ethical concerns. "
While some of these deeper criticisms may be true
in part or for some organizations, we think a more
multifaceted and charitable interpretation [15], [16]
is both appropriate and likely to be beneficial toward
seeking positive change. Indeed, significant literature
has challenged the notion that firms or individuals
within firms are characterized by purely " rational "
economic behavior [17], [18]. We argue instead that
organizations are best understood not as monolithic
single actors, but as multiple coordinating and competing
coalitions of individuals [19]. Individuals within a
single organization may have multiple and conflicting
preferences and roles. Organizational motives should
therefore be considered a complex composition of
genuine ethical concern, economic logic, signaling
and framing strategies, and efforts to promote change
both internally (i.e., within the firm) and externally
(e.g., in regulation and public opinion) [20].
Regardless, the tension between economic and
pro-social motives, and the incentive structure that
this produces should be taken seriously. A key question
then is what policies and practices can shape
and enable firms to weigh competing values in a
responsible way. There are many possible pathways
here, both internal and external to organizations:
industry collective self-regulation and formal governmental
regulation, shifts in public and societal
norms, changes to education and training of engineers,
and adoption of a broader focus on stakeholder
benefits along the lines on B Corps, benefit
corporations, and a triple bottom line orientation.
Organizations that already have a strong commitment
to social good can also consider " value levers "
[21] that facilitate responsible practice, as well as
some of the recommendations we suggest below.
The complexity of AI's impacts
A second explanation for the principles-to-practices
gap is that AI's impacts on well-being-positive
June 2021
or negative-are more complex than is sometimes
assumed. Site-based research has identified that
engineers are often focused on single products
and the physical harm they may cause rather than
broader kinds of social, emotional, or economic
harms [22]. Even as conversations surrounding
responsible AI increase, most work centers around
a relatively small subset of issues, most often bias
[23] and transparency [24] in particular AI models.
This approach involves exposing and then
attempting to mitigate bias in algorithms as well as
trying to improve interpretability or explainability
given the black-boxed nature of certain AI models
which can make decision-making processes and
their outcomes opaque. Other commonly emphasized
issues include privacy, reliability, and safety.
Yet, these prominent issues most familiar to engineers
and dominant in AI ethics research constitute
only a subset of social and ethical risks and impacts
related to AI. In place of such a narrow focus, AI
should be understood to impact a wide variety of
aspects of human and societal well-being, such as
human rights, inequality, human-human relationships,
social and political cohesion, psychological
health, and more. AI can also impact natural ecosystems
and animal life, impacts that are arguably both
intrinsically and instrumentally important.
Moreover, many of these harms do not arise in
a straightforward way from a single AI product, but
from many AI systems influencing human social and
economic life together over time. For example, algorithms
on social media designed to steer consumers
to entertaining video clips have also led to so-called
filter bubbles that may foster political polarization,
misinformation and propaganda, targeting of minority
groups, and election interference [25]. AI as
instantiated in autonomous vehicles has potentially
massive implications for physical infrastructure,
energy and environment, traffic fatalities, work productivity,
urban design, and unemployment [26].
In short, addressing AI principles in full seriousness
requires an expansive scope of attention to the
full set of issues influencing human well-being. This
requires looking well beyond a narrow set of topics
such as bias, transparency, privacy, or safety and
treating them as independent issues. Instead, the full
range of topics and their complex interdependencies
need to be understood. However, such a task
can be enormously difficult.
83
IEEE Technology and Society Magazine - June 2021

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - June 2021
