Editorials/Opinions Analysis For UPSC 10 June 2023
- Open Network for Digital Commerce (ONDC)
- India’s AI bias: a danger to equity and equality
Open Network for Digital Commerce (ONDC)
The Open Network for Digital Commerce (ONDC), which aims to “democratise e-commerce” and “provide alternatives to proprietary e-commerce sites,” is expected to be formally launched by the Union government this year. Although it has invited businesses to sign up for the ONDC platform, well-known e-commerce giants like Amazon and Flipkart have been hesitant to do so.
GS Paper- 3 – Science and technology – Latest technological development
What does ONDC mean to you? How would it open the door for more enduring e-commerce? (150 Words)
The existing “platform-centric model to an open-network model” of the e-commerce business is what the government intends to change. The Unified Payments Interface (UPI) project, which is widely regarded as a success, served as a model for the ONDC. No matter which payment platform a person is registered on, the UPI initiative enables them to send or receive money. In a similar vein, the government aims to guarantee that participants in the e-commerce market can conduct transactions independent of the platforms on which they are registered. Therefore, a buyer registered on Amazon, for instance, may directly buy goods from a vendor who sells on Flipkart under the terms of ONDC. The government has mandated that businesses identify themselves on the ONDC in order to make such transactions a reality. Thousands of retailers have already joined the site since the pilot edition of ONDC started last year in a few significant cities. However, Amazon and Flipkart have not yet integrated their primary shopping sites with the ONDC network.
Why is it being promoted by the government?
The administration is certain that the ONDC will end the dominance of a few major platforms in the e-commerce business. According to this, private platforms now control and operate the “silos” that make up the e-commerce market. For example, it has been claimed that Amazon and Flipkart are promoting specific seller firms in which they have indirect ownership stakes. Swiggy and Zomato, two food delivery apps, have recently come under fire for allegedly charging vendors exorbitant commission fees. The government wants to level the playing field and eliminate the need for private platforms with an open network like ONDC that links buyers and sellers across platforms.
What claim critics?
Critics claim that the alleged advantages of an open network for digital commerce are currently far from certain. For starters, given the current platform-centric e-commerce model, vendors are already free to offer their products across numerous e-commerce platforms. Buyers frequently browse several platforms. Additionally, there are services like price comparison that are provided by numerous private websites, bridging the information gap and assisting consumers in making better choices. Critics contend that platforms like Amazon and Flipkart may not actually have a captive audience of buyers and sellers, which would explain their dominance of the online marketplace. Furthermore, it’s possible that the purported monopoly that platforms are believed to hold is identical to the restricted monopoly that any company in the modern era holds on its assets.
The next step
As the government implements the ONDC, the ability of the technocrats to develop an effective substitute for e-commerce platforms that can function seamlessly will be put to the test. It’s unclear whether and how the government’s open network will list goods provided by different suppliers. E-commerce platforms are typically compelled by competition to prominently list the goods that are most likely to appeal to customers. The ability of sellers to complete customer orders has a big impact on how merchants are on-boarded and listed. Platforms may spend money developing unique on-boarding and listing procedures. Platforms may stop making such investments if the open network’s regulations forbid them from reaping the rewards. This will eventually have an impact on the standard of services offered to customers. The biggest hurdle for ONDC may be creating a successful marketplace for the exchange of goods and services.
India’s AI bias: a danger to equity and equality
- Artificial intelligence (AI) is a fast evolving field of study that has the potential to drastically alter a variety of facets of our daily life.
- The possibility that AI could be used to reinforce current prejudice and discrimination is an increasing worry, though.
- There has been an increase in warnings against artificial intelligence (AI) from technology leaders, researchers, and experts, including Geoffrey Hinton, a pioneer in artificial intelligence, Elon Musk, a co-founder and CEO of Tesla, Emad Mostaque, a British AI expert, Cathy O’Neil, an American mathematician, data scientist, and author, and Stuart J. Russell, a British computer scientist.
- The World Health Organisation recently advised caution when utilising AI in public health care.
GS 3 : Science and Technology, GS 4 : Ethics
Artificial intelligence: What is it? In view of the “NITI Aayog initiative for Responsible AI,” what AI should be implemented in India? (150 Words)
What precisely is artificial intelligence?
The replication of human intelligence functions by machines, particularly computer systems, is known as artificial intelligence. Expert systems, natural language processing, speech recognition, and machine vision are some examples of specific AI applications.
How Can AI Bias Occur?
- AI algorithms have the ability to learn from the past and predict the future. If the data they are trained on is biassed, this implies that they may be able to maintain current prejudices and discrimination.For instance, computers may be able to sort through large amounts of resumes, career histories, and past employee performance in an organisation and discover that men are generally more productive than women.
- According to a recent research by the United Nations Development Programme, the financial services, healthcare, retail, and gig worker industries are most affected by algorithmic prejudice. This survey also noted that, despite having equal financial backgrounds, AI-generated credit scoring had a propensity to give women worse scores than men. Since the majority of the data used to train the model came from China and the United States, with little representation of other communities even within these two nations, research on applications in health-care diagnoses has found severe biases against persons of colour.
- These algorithms lack a moral compass and, unlike people, do not challenge stereotypes, conventions, culture, or traditions, despite the fact that they are programmed to get better through feedback loops and programming changes. Machines lack the sense of justice and humanity that society relies on, particularly when it comes to minorities and underrepresented communities. Additionally, computers apply learned patterns to the entire population without understanding if the data used to train them is complete, diverse, and adequately representative of all communities and groups that may be impacted by their use.
- Given the history of the Indian States and their diverse languages, ethnicities, cultures, and traditions, this threat has the potential to increase inequalities and marginalised groups’ exclusion, which could result in high costs for their livelihoods, opportunities, well-being, and quality of life.
Others Risks related to AI
- Superintelligence: Its intelligence would exponentially grow during an explosion of intelligence and could significantly outperform humans.
- Technological unemployment: As machines become capable of completing jobs that are currently performed by people, it is possible that AI will result in job displacement. Increased unemployment and societal instability may result from this.
- Terrorists, criminals, and rogue nations may deploy different types of weaponized AI, such as lethal autonomous weaponry and advanced digital warfare. Over fifty nations were said to be studying battlefield robots by 2015.
- Privacy issues: AI systems have the potential to be used to gather and examine a lot of personal data about individuals, which could lead to privacy abuses.
How Can AI Be More Fair?
- To create algorithms and models that help our citizens, we need public data that is clear, organised, digitalized, and subject to good governance. Before adopting this breakthrough, business and government must take prudence and make significant investments in its research, development, and evaluation.
- Despite the fact that our administration and services are able to operate at a large scale, the problem of the available data volumes can be solved, it is crucial that we give the development of AI the attention it deserves. Consider the NITI Aayog’s Responsible AI project.
- It’s crucial to guarantee that the population-representativeness of the data used to train AI systems is preserved. This entails incorporating information from people from all backgrounds, especially members of underrepresented groups.
- Governments should collaborate with the private sector and professionals. It is essential to stress that the objective must be to create inclusive, diverse, and heterogeneous intelligent machines.
- The implementation of this innovation should complement rather than impede our efforts to bring about positive change and equality and equity for all.
- India’s use of AI has the potential to be a significant force for social progress. However, it is crucial to be conscious of the possibility of AI bias and to take precautions to lessen this danger.
- We can make sure that AI is used to benefit every member of society, regardless of their background, by taking these actions.