We need to act on AI now, not have an act for it
When Bangladesh embarked on its journey towards Digital Bangladesh in 2009, many were sceptical about it. But as time progressed, we all saw how the vision started to become a reality.
This vision, at its core, aspires to create a nation that is adept at solving problems at all spheres of life through innovative application of digital technologies. The government has made it abundantly clear that Artificial Intelligence (AI) is going to play a pivotal role in implementing the Smart Bangladesh vision. Following this vision, the government has recently unveiled a draft National Artificial Intelligence (AI) Policy 2024 for public consultation.
For all latest news, follow The Daily Star's Google News channel. There is a good reason why the government has decided to use AI as the fulcrum to realise the goal of Smart Bangladesh. Unlike other digital technologies, the potential of application of AI is literally all around us. Starting from our personal lives, to modernising public service delivery, the scope for AI is limitless.
Be it public transport or AI-driven personal vehicles, personal healthcare solutions or the public healthcare system, from individual human resource productivity or national competitiveness in productivity levels, every imaginable aspect of our individual, societal, as well as national issues can have a transformative impact if we can smartly apply AI to solve our problems.
But the question is this: how do we facilitate AI to deliver the dividends for us? If we look around, we can see that every country in the world is trying to strike a balance between innovation and regulatory oversight. There is palpable consensus on adopting more of a business-friendly approach to AI regulation, by avoiding excessive restrictions.
The government has been trying to create a pathway for AI in Bangladesh by preparing the National Strategy for AI in 2020, followed by the recent release of the draft AI Policy in 2024. Having read the draft policy on AI, I felt that it provides an excellent template to foment the use of AI in every sector. The institutional framework outlined in the policy to pursue AI projects is well thought through. On top of that, the sectoral plans for application of AI provides an excellent starting point.
But what puzzles me is the stated desire of the government to introduce an Act for AI. When we are supposed to allow as much room as possible for our AI practitioners to fully demonstrate their talent, we are planning to limit what they can and can't do along with defined punitive measures through the AI Act. I am certain that this is not how you invite people into the fold of new technology.
As of now, the European Union (EU) is the only entity to have enacted an AI Act. At the heart of the Act, it is mandatory to ensure that AI platforms are monitored or overseen by human beings, not another AI platform. It's worth noting that many AI experts have termed this as a knee-jerk reaction as they consider a law on AI to be too premature at this stage.
The US does not have a federal law covering AI, nor is there any universal definition for AI. It is currently governed by a mix of decentralised existing federal and state legislations, industry itself and the courts. Through an executive order last year, every US government agency was tasked to set up working groups to evaluate AI, develop regulations and establish public-private engagement.
In United Kingdom (UK), the government has unveiled its response to AI Regulation White Paper consultation in February 2024. They don't have any plans to codify that into law for now. It advocates a context-sensitive, balanced approach, using existing sector-specific laws for AI guidance.
In India, the upcoming Digital India Act is set to focus on the regulation of high-risk AI applications. No plan to enact separate legislation is afoot. Singapore also doesn't have any AI legislation; they have a sector specific approach to overall governance and regulation. Japan also has a relatively hands-off approach and has been encouraging AI development and application across various sectors.
The Association of South East Asian Nations (ASEAN) has issued a guide to AI governance and ethics in February 2024. The national-level recommendations include nurturing AI talent, upskilling workforces and investing in AI research and development. Australia also doesn't have any AI legislation; the government there is approaching it with a voluntary ethics framework.
It's worth noting that the core purpose of having a law is to create a framework for dos and don'ts in a particular area with the option to resort to the legal system to settle disputes or punish offenders of the law. The question here is, how do we know what is doable and what is not, when we don't have any prior experience with AI in Bangladesh.
Even if we consider enacting a law, we need to ascertain areas where government regulation is needed, in light of the global best practices. AI law or policy considerations should include the use and processing of personal data, privacy, infringement, surveillance, algorithm bias in customer interactions, data sovereignty, monitoring AI based platforms, cybersecurity, and social norms and values etc. Most importantly, we need to focus on the fundamental ethical aspects of AI, which are more universally agreed upon compared to specific AI regulations.
We must realise innovation involves a very messy and unstructured process. The key to innovation is to have a creative mindset that can go beyond conventional thinking to come up with the simplest of solutions to complex problems. Putting barriers on this through an AI Act is the last thing we need at this moment.
If we want to meet the export earnings target of $5 billion from the ICT sector, we need to facilitate our developers to catch up with the rapid pace of AI development globally, instead of scaring them off with an act that comes with punitive measures. More AI regulation risks stifling new start-ups who lack the resources of the globally dominant platforms. We need to focus on creating a large pool of highly skilled human resources in AI. The draft AI policy provides a baseline to embark on this AI journey.
Shahed Alam is a barrister and telecom expert.
Views expressed in this article are the author's own.
Follow The Daily Star Opinion on Facebook for the latest opinions, commentaries and analyses by experts and professionals. To contribute your article or letter to The Daily Star Opinion, see our guidelines for submission.
Comments