AI Regulation: Balancing Innovation and Control

 The digital revolution has transformed the complete working and performance of the industries enhancing the business across several dimensions. All the aspects of communication, customer services, business development, finance management, workforce management, business planning followed by research and development; have been optimized and enhanced with the application of the technologies. The developments in the software applications systematized the business operations through synchronized software tools. The further advancements led to the new revolution of Artificial Intelligence (AI), Machine Learning (ML) and Internet of Things(IoT).

The AI systems have helped in executing the repetitive tasks with accuracy, speed and reliability. The basic communication in customer service department is now taken care by the AI chatbots that have built-in system that answers the users questions and engages the customers on a professional level. The involvement of AI is now in every field of communications, banking, education, entertainment, healthcare, business and more.

As AI technology became popular and commonly adopted, it changed the course of human involvement in certain functions. It started assisting the users with all sorts of solutions like answering the questions to performing specific functions. Soon there came situations reflecting the wrong use of AI in acts of cheating, cyberthefts, frauds, etc. There came an urgent need for AI regulation.

Government organizations along with several leading technology organizations have decided to formulate AI policy that can standardize and regulate the AI usage. The objective of having a sound AI policy is to bring all the AI activities under unified AI control that will support in building and maintaining the ethics, safeguard interests of the users and create a world of cooperation and support.

The AI regulations have been classified in three categories of accountability and responsibility of AI systems, governance of artificial intelligence systems and managing privacy with safety concerns.

The need to ensure long-term beneficial AI is known as the AI control problem. Other social responses, like doing nothing or outlawing AI, are viewed as impractical, while approaches like augmenting human capabilities through transhumanism techniques like brain-computer interfaces are seen as potentially complementary. Regulation of AI can be seen as a positive social means to manage this problem. Regulation of AGI research centers on the function of review boards at all levels, from corporate to university, and on promoting research into AI safety along with the potential for differentiated intellectual progress (giving protective over risky strategies in AI development) or conducting global mass surveillance to carry out AGI arms control.

The ‘AGI Nanny’ is a proposed human-controlled strategy that aims to deter the development of a dangerous superintelligence and tackle other significant threats to human welfare, like the disruption of the global financial system, until a safe and true superintelligence can be created. To monitor and safeguard humanity from harm, it involves building an artificial general intelligence (AGI) system that is more intelligent than humans but not superintelligent. This system is then connected to a vast surveillance network.

Regulation of ethically aware, conscious AGIs focuses on integrating them into human civilization; this can be broken down into two categories: legal issues and moral rights.

Regulation of AI has been viewed as onerous, potentially impeding the advancement of artificial general intelligence. Launched in June 2020, the Global Partnership on Artificial Intelligence (GPAI) emphasized that, in line with the OECD Principles on Artificial Intelligence (2019), AI development must respect democratic values and human rights in order to maintain public confidence in the technology.

Australia, Canada, the European Union, France, Germany, India, Italy, Japan, the Republic of Korea, Mexico, New Zealand, Singapore, Slovenia, the United States, and the United Kingdom are the fifteen founding members of the Global Partnership on Artificial Intelligence. The GPAI has 29 members as of 2023. The OECD is located in Paris, France, and hosts the GPAI Secretariat.

The International Centre of Expertise in Montréal for the Advancement of Artificial Intelligence supports two of the four areas covered by GPAI’s mandate: data governance and responsible AI.

In May 2019 the OECD AI Principles and in June 2019 the G20 AI Principles were ratified.Ten ‘AI Government Procurement Guidelines’ were released by the World Economic Forum in September 2019.The European Union released its draft AI promotion and regulation strategy paper in February 2020. The UNCRI Centre for AI and Robotics is one of the organizations inside the UN that has started to advocate for and conduct discussions about various elements of AI regulation and policy.

UNICRI’s Center published two reports in collaboration with INTERPOL: “AI and Robotics for Law Enforcement” in April 2019 and “Towards Responsible AI Innovation” as a follow-up in May 2020. In November 2019, UNESCO announced the start of a two-year process to create a “global standard-setting instrument on ethics of artificial intelligence” during its 40th Scientific Session. Stakeholder opinions were gathered through UNESCO forums and conferences on AI in order to achieve this goal. Legislative gaps should be remedied, according to a draft of the UNESCO Ad Hoc Expert Group’s recommendation on the ethics of AI, which was released in September 2020.

At its General Conference in November 2021, UNESCO presented the international instrument on the ethics of AI for adoption; it was eventually adopted. Although the UN is making strides in managing AI globally, its institutional and legal capacity to handle the existential risk of AGI is more constrained.

The leading AI companies are coming together to form a committee for independent board for AI oversight to deal with the issues related security processes.