The race to regulate AI

Recent developments in Artificial Intelligence have advanced our digital capability, further enhancing our ability to learn and express knowledge. AI enables computers and digital devices to perform tasks such as learning, reasoning and problem-solving that are typically solved by human intelligence.

Recently, the development of apps like ChatGPT has attracted a renewed interest in AI globally. With its ability to create new and unique content, such as text, images and code, using learnings from wider datasets, AI is poised to revolutionise the world in a whole new way. 

At the same time, it is crucial to ensure that AI is deployed and used responsibly while addressing bias, ethical and privacy concerns, amongst others. 

Recently, Law Minister Anisul Huq announced the drafting of the Artificial Intelligence Act (AI Act) by September 2024. Meanwhile, the Ministry of Posts, Telecommunications and Information Technology has released the Draft National Artificial Intelligence Policy 2024 (Draft Policy). The draft has already sparked discussions among AI enthusiasts eager to understand the potential provisions of the upcoming law and its impact. 

When it comes to enacting an AI Policy or an AI Act, Bangladesh must consider the big gap between developed and developing countries, and use the global benchmark to turn their policies into action. AI technologies are evolving at an exponential pace. Advances in AI models, especially large language models and generative AI, are revolutionising the innovation space, starting from smart agriculture solutions to modelling climate change. 

Bangladesh has traditionally looked to India to draw on its legislative framework on different laws due to our shared socio-economic similarities. In that respect, Bangladesh will find itself charting a new territory since India does not have a comprehensive legal framework on AI yet, and is operating based on AI advisories issued from time to time. 

The enforcement mechanisms are weaker and resources are limited in the Global South as opposed to the Global North, which often leads to policies based on guesswork or political interests. Bangladesh should instead consider analysing the existing AI regulations worldwide, such as the AI Act of the European Union, the UAE National Strategy on AI 2031, the ASEAN Guide on AI Governance and Ethics, the UN Resolution on AI, etc. 

The European Union enacted the world's first comprehensive AI law in March 2024. Aimed at protecting fundamental rights, democracy, the rule of law and environmental sustainability from high-risk AI, the Act seeks to boost innovation and establish Europe as a leader in the field. 

The Act in particular bans certain applications of AI that threaten citizens' rights, including biometric categorisation systems based on sensitive characteristics and untargeted scraping of facial images from various sources. 

In principle, the law also prohibits the use of biometric identification systems (RBI) by law enforcement agencies. It further provides clear obligations and transparency requirements in relation to high-risk uses in critical infrastructure, education and vocational training, employment, essential private and public services, etc. It also provides measures to support innovation and SMEs. 

However, the Act, while hailed for its efficiency, has faced criticism related to privacy concerns, from human rights agencies. For instance, there are concerns that Article 19 fails to completely ban the use of emotional recognition technologies and real-time remote biometric identification in publicly accessible spaces.

The Middle East has yet to see the publication of an overarching dedicated AI law in the region. The UAE issued a comprehensive national strategy on AI in November 2023, setting out eight key strategic AI objectives in education, healthcare, transportation, government services, etc for the period up to 2031. The strategy seeks to establish the UAE as a pioneer in AI research development. 

The Association of Southeast Asian Nations (ASEAN) released its Guide on AI Governance and Ethics in early February, mapping out a voluntary and light-touch vision of how national governments in the region can constrain the worst aspects of AI, yet profit economically from its advance, making it a business-friendly approach to AI regulation. 

In a more recent development, the UN recognised that AI can help accelerate the achievement of the 17 Sustainable Development Goals, and it stresses the urgency of "achieving global consensus on safe, secure, and trustworthy artificial intelligence systems." 

This development is the first attempt to bring all nations on common ground regarding AI regulation and reinforces the importance of propagating ethical and safe use of AI in every nation's national strategies but not "without adequate safeguards or in a manner inconsistent with international law." AI regulation and AI policies are welcomed if they can ensure it is "safe, secure, and trustworthy" AI.

Hong Kong and Singapore are examples of countries taking a pro-innovation, principles-based approach. A notable exception is China, which has taken a prescriptive approach and introduced three AI-specific regulations to regulate specific AI technologies (including, most recently, generative AI). 

Although other countries in Asia have started discussing whether AI-specific laws should be introduced (including India, Japan, South Korea and Taiwan), these discussions have not yet crystallised into binding laws or regulations.

Recently, The French Competition Authority imposed a 250 million Euro fine against Google for not complying with previous commitments and using news articles to train its AI system (Bard/Gemini). The United States Securities and Exchange Commission (SEC), has charged investment advisers for making false and misleading statements about their use of AI, as AI washing could hurt investors. 

YouTube also requires creators to disclose when 'realistic' content is made using generative AI. The US state of Tennessee enacted the ELVIS Act (Ensuring Likeness Voice and Image Security Act), protecting artistes from AI deep fakes, which is set to come into effect from 1 July 2024. 

The US government agencies will be hiring Chief AI Officers (CAIOs) who will be responsible for (i) Coordinating Agency Use of AI, (ii) Promoting AI Innovation and (iii) Managing Risks from the Use of AI. 

With our aim of achieving the Smart Bangladesh Vision 2024, a problem-specific AI solution that keeps a balance between economic growth and upholding human rights and democratic values would be ideal. 

Barrister Tasnuva Shelley is an advocate at the Supreme Court of Bangladesh.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.