AI’s widespread deployment is prompting companies to explore how best to ensure that it remains fair and free from bias.
Examples include bias in AI tools that determine who receives loans, jobs or healthcare prioritization. Another problem lies with unclear data-sharing policies.
Ethics of Robotics
One of the central ethical questions surrounding robots involves whether or not to grant them moral status and rights. Some believe this right should not be granted because robots are machines without intrinsic human characteristics (Bryson 2010), while others argue that as more intelligent robots emerge they might develop an internal sense of value which gives them their status and rights, regardless of whether we share those beliefs (Gordon 2020a).
Most discussions regarding robots’ military application have focused on military use alone. Some individuals have voiced concerns over potential “killer robots”, and there has been a campaign against their development; this discussion, however, does not take into account wider ethical considerations involved with using more robots in warzones.
Some individuals have attempted to address these problems by creating robots they think will act ethically, such as Guarini’s (2006) system in which a neural network was given known correct answers and then asked to solve new ethical dilemmas independently. Unfortunately, his approach proved ineffective because the robot couldn’t adequately reflect upon and represent all circumstances surrounding each ethical decision made.
Other approaches have attempted to develop robots capable of making moral decisions by providing it with a set of values that should guide its behavior, yet this approach proves challenging since it’s hard to identify exactly which values a machine should uphold and how those will influence its behavior in any given circumstance. Furthermore, external factors may influence it into making biased choices (Suddart 2015).
More practical solutions may involve broadening existing constraints on technology design. For instance, rules exist requiring manufacturers to ensure product safety and avoid false advertising; another possibility would be creating a code of ethics for AI engineers similar to what exists for medical doctors.
Ethics of Artificial Intelligence
Artificial Intelligence (AI) is rapidly altering our world. Already embedded into numerous aspects of daily life – from YouTube recommendations and computer-generated music and visual effects in movies and video games, to medical research for faster drug discovery, autonomous vehicle safety improvements, and autonomous driving safety enhancement – AI has already made itself felt in our daily lives. But as AI becomes ever-more pervasive and sophisticated, new ethical considerations need to be considered when planning its deployment.
One major concern of AI systems is their potential to make decisions with profound effects on people’s lives, such as who gets a loan, admission into university programs or product recommendations. Therefore, these systems should be transparent about their decisions by explaining and justifying their reasoning (often called “explainability”) – seen as an essential ethical standard (Floridi et al. 2018).
Concerns related to AI include using collected data collected by AI to target and manipulate individuals online or offline using nudges, manipulation or deception. This can be accomplished using machine learning algorithms that collect, interpret and deliver targeted messages or products (McCormack 2019). Furthermore, their use for surveillance could pose ethical concerns that lead to loss of privacy (McCormack 2019).
AI development uses considerable energy, leading some researchers to be concerned that its deployment could have an adverse impact on the environment (Haggstrom 2016; Ord 2020). This energy gap issue is sometimes known as “research vs deployment gap” of AI development (Haggstrom 2016; Ord 2020).
Finally, ethical concerns exist concerning whether machines can have morality and personhood. This debate often revolves around Kantian tradition in ethics which holds that greater levels of rationality don’t always translate to greater levels of morality or the capacity for acting morally.
As technology develops and is adopted more widely, ethics guidelines become ever more essential to helping prevent existential threats to humanity and ensure we build AI systems which respect our ethical standards. Such guidelines could involve legal or regulatory approaches as well as codes of conduct or self-regulatory mechanisms.
Ethics of Cybersecurity
As cybersecurity professionals deal with sensitive data and powerful tools that could potentially cause harm or impact people, they should be cognizant of potential ethical concerns. They must adhere to set ethical standards while understanding game rules.
Cybersecurity professionals must adhere to principles such as privacy, beneficence, justice and legal compliance when conducting professional activities as well as research that involves people they work with. Furthermore, they should respect individuals’ autonomy by refraining from coercive tactics that undermine autonomous rational choice; this principle applies both professionally and when conducting research using data derived from people working together.
cybersecurity researchers are concerned with the use of AI systems for surveillance and monitoring, which could compromise users’ privacy rights. Researchers fear AI-based surveillance systems will be exploited for targeted advertising, discrimination and profiling based on demographics or other criteria; also fearing these algorithms being misused to manipulate behavior online and offline and undermine autonomous rational choice; attempts at manipulating behavior have long existed; however they will take on greater significance with AI technologies being able to identify and target specific individuals more readily than ever.
AI poses another ethical quandary when used to enhance or replace human performance, including replacement. Some technophiles such as Kurzweil and Bostrom support “transhumanism,” in which humans survive by adopting alternative physical forms or uploading themselves onto computers (see Human enhancement).
AI also raises ethical concerns. For instance, its use in training data could result in it having biases which lead to unfair or inequal results for groups of people – for instance a cybersecurity system could identify software used disproportionately by certain groups as malicious software and flag it accordingly.
Ethics are at the core of any profession, including cybersecurity. Physicians, attorneys, and other professionals abide by clear ethical codes with penalties in place if violations occur. Cybersecurity professionals must likewise follow these regulations with strict observance – it’s crucial they be in place and that their awareness remains high.
Ethics of Big Data
Big data ethics is an emerging area of study designed to address ethical considerations related to the collection, storage, analysis and publication of large sets of information. While information ethics focuses on legal and privacy concerns of librarians and information professionals; big data ethics more directly addresses data brokers or large organizations which collect and analyze structured or unstructured data sets.
Big data has an array of applications in healthcare, business, social media and scientific research. As this technology becomes more widespread its ethical implications can become complex; including concerns related to privacy, fairness and transparency as well as making sure any research conducted with big data does not infringe upon individual rights or compromise scientific integrity or public wellbeing principles.
Big data may also raise concerns of discriminatory use or breaching trust between businesses and people, so it is vital that businesses establish and adhere to an effective privacy program, disclosing all customer data being gathered as well as using reliable third-party partners when handling sensitive data.
Artificial intelligence may produce machines with consciousness. Many researchers speculate that, should AI continue its current pace of improvement, it may eventually be capable of experiencing emotions and making independent decisions without human interference – raising questions as to whether these machines have rights similar to that of living organisms or are eligible to be treated like such beings.
Other ethical considerations surrounding data mining include its potential to reveal sensitive personal information about individuals, such as medical histories or financial details, which could then be used for fraudulent or criminal purposes such as insurance fraud or identity theft. There are also ethical concerns around using big data for medical or behavioral research; specifically its use for “human enhancement,” with advocates seeing it both positively (Kurzweil) and negatively (Bostrom). This trend towards transhumanism includes both those who see its advantages (Kurzweil) as well as potential dangers (Bostrom).
