Artificial intelligence is a relatively new technology that emerged in the twenty-first century. The simulation of human intellectual processes by machines, particularly computer systems, is referred to as this notion. Learning (the acquisition of knowledge and the rules for applying it), reasoning (the application of rules to arrive at approximate or definite conclusions), and self-correction are examples of these processes. Expert systems, speech recognition, and machine vision are examples of AI applications. Artificial intelligence is being more widely employed, and it can now be found in almost every social media and internet platform.[1]

This artificial intelligence system may be claimed to track each and every individual's movement. Each of us is exposed to technology and its negative effects; unfortunately, the dangers of this technology are unknown to the general public, and it may be a major source of violation of the right to privacy, since nothing is private in this era of artificial intelligence. This study will describe the extent to which artificial intelligence has intruded into our lives, as well as the legal safeguards in place to limit artificial intelligence's usage. This article will also offer a solution to the problem and an analytical examination of the situation.

Relation between Artificial Intelligence and Social Media

We are primarily worried with social media, and when we talk about social media, we must not forget about its backbone, namely the database and its technology. Artificial Intelligence is the primary technology that powers today's social media networks, and the use of AI databases is on the rise. When a person enters into a social networking site, a new database is created in its name, and data begins to accumulate, which quickly becomes public.

One of the platforms that is being rapidly impacted by AI development is social media. In the last decade, social media has become a key platform for digital marketing, and it has just made a significant entry into the consumer journey. With roughly 3 billion individuals daily checking into social media sites, AI has played a significant role in the process of improving the social media sites' interfaces. AI may be thought of as the social media's brain, providing direction to marketers and assisting them in developing more successful customer-targeting strategies.

Let's take a look at how the main social media companies, also known as social media marketers, are using AI to better the consumer journey online:
Social Listening
Customers that shop online leave a large digital footprint, and the database collects a lot of information on what the client wants, what their interests are, and what they're chatting about with each other. It maintains track of each and every conversation the consumer has. There is a distinction to be made between social monitoring and social listening. Social monitoring maintains track of whether customers are talking about a certain brand or topic of interest; it essentially records keywords. This approach is elevated to a new level using social listening. Social listening, which employs AI algorithms, discovers important phrases, concepts, and brands that are highly relevant to the firm.

Customer Segmentation
It is a digital marketer's most significant task; it is the process of classifying customers based on their age, area, online presence, content preferences, previous purchases, and other factors so that it is much simpler to put the correct product in front of the right client. Collecting data and analysing it according to groups was a challenging process in the past, but today there is AI that is much more powerful and has a solid technical basis that aids in this process. Artificial intelligence has been educated and primed in a highly effective way to sift through such large numbers, clicks, and preferences in order to organise such a large quantity of data.

Unique customer experience
Artificial intelligence has a lot of potential in the future, and it may be ingeniously integrated to social media to benefit users without invading their privacy. It is critical to look into the misuse of AI because it is a user's most essential right. There is a lot of opportunity for AI advancement in terms of database setup for a better user experience.[2]

The artificial intelligence (AI) utilised by the biggest social networking sites is quite advanced, and it makes the user experience very pleasant. The giants' database is built up in such a manner that it provides the user with a customised interface and the experience that they anticipate from the site. Interests, purchases, and anything a user looks for on the internet are all taken into consideration, and the AI builds the interface accordingly. Because of the AI that is set up in the interface, one may always see all posts and adverts linked to his interest on social networking sites. The AI maintains track of the user's actions and then distributes it to the appropriate advertisement firm.[3]

The impact of Artificial Intelligence on Social Media

Artificial intelligence (AI) has advanced at a fast pace, from the introduction of SIRI to self-driving automobiles. Old school science has long depicted AI as robots with human-like features; however, AI is far broader, encompassing everything from Google's search engines to IBM's Watson to autonomous weaponry. One of the most crucial areas in the fast-growing AI competition is social media platforms that use AI in their databases.

Today's artificial intelligence is referred to as narrow AI (or weak AI) since it is built to execute a certain purpose (e.g. only facial recognition or only internet searches or only driving a car). Many researchers' long-term objective, however, is to develop generic AI (AGI or strong AI). While narrow AI may surpass humans in a single skill, such as playing chess or solving math problems, AGI would exceed humans in practically every cognitive endeavour.[4]

'So, how does Artificial Intelligence work in understanding the human behavior on the three most popular social networking sites � Twitter, Facebook and Instagram?'

Artificial Intelligence in Facebook, Twitter and Instagram

Facebook's business model is based on learning about its user's behaviour and building businesses around it. Facebook has approximately 2.38 billion monthly users who share around 4.75 billion pieces of information every day, making it tough for data analysts to analyse such a massive volume of unstructured data. Instagram is one of the most popular applications nowadays, and it is mostly utilised by teenagers. In the year 2010, it swept over social media like a storm as a social networking site. It claims to have a user base of one billion people as of 2019. However, as it has grown so much, it has given place to a plethora of advertising methods such as brands and the like. According to statistics from 2016, over 98 percent of fashion firms have an Instagram account.

Twitter is another social networking site that has one of the most widespread user base of about 328 million active users per month. Apart from the AI uses by Facebook, Twitter uses some more AI that is event promotion, marketing, etc.

• Deep Learning
  Deep Learning is a type of technology that can help in this situation because it doesn't require any specific data from an image. For example, many of the same kind of photographs and videos are posted throughout the social networking site. Deep learning is capable of recognising the content of specific films and images. It has more meta and text-based techniques to analyse that stuff. It considers the look of the items as well as the photographs and videos. This aids in the selection of the appropriate location for the appropriate individual.
   
• Deep Text
  Deep Text is a kind of tool developed by Facebook team which helps in analysing the posts of the users and focus on the words in them. There is a way known as neural networking that determines the underlying meaning of the text that is present in the post. There is no specific database which can be referred instead it uses its own algorithms to understand the meaning of the text in the post. Facebook itself analyses words across 20 different languages, Deep Text assigns certain labels to each word.
   
• Face Recognition
  Deep Face is another technology that is used by Facebook that recognises human faces on the images and it can recognise faces in two or more than two different faces. Moreover, Facebook claims that this feature has a lot of efficiency and can even be more accurate than the human itself can recognise the face � the success rate in recognising the faces is 97% and 96%.
   
• Tweet Recommendations
  Twitter uses Artificial Intelligence while recommending the tweets to the users on their timeline. With about 5,787 tweets per second as per the data of year 2014, the tweets that the users see are most relevant to their interests. The process that is used by Twitter is known as Natural Language Processing (NLP), which helps in analysing the mammoth number of tweets that are posted per second. NLP is used to provide insights to the users on the basis of their preferences, so that they get to see the tweets which they are more interested in.
   
• Discovering experiences
  The search function of the Instagram application is very much dependent on the photos that are being uploaded on Instagram on a daily basis. With approximately 70 million photos are uploaded everyday on this platform, these posts constitute to be a huge database by using hashtags, and trending information. The Artificial Intelligence used by Instagram helps the users to see pictures according to their interests, activities, experiences, events, eateries, etc.
   
• Instagram's Own Artificial Intelligence Influencer
  "Miquela" is the name of this AI Influencer which is hard to believe that it is a "fake" account. It is just like another influencer account, it has around 1.6 million followers giving tips on fashion blogging It is said to be that it is a human robaot that it posts photos and videos and captions them on its own, alike a fashion model posting stuff. Even it was collaborated with Prada in February 2018.

AI and social media are inextricably interwoven, with AI serving as the social media's backbone. Both are proven to be quite useful to the company. Artificial Intelligence services may be used by these organisations and marketing managers by researching user behaviour and searching patterns. This allows them to eventually reach their intended audience. Nowadays, commerce on these social networking sites is booming at a rapid pace. Many small businesses have begun on Instagram by creating their own page and advertising and selling there.[5]

"By studying buying behaviours, people's everyday circumstances, and what inspires them to make decisions, marketers may customise information for their consumers." And AI can pick up on these tendencies far faster than humans.

Companies are finding it more difficult to stay on top of trends in the ever-changing social media world. That is why they are turning to AI, which is a wise decision. Machine learning algorithms evaluate everything that happens on social media in real time and turn it into concrete data that marketers can use.

Data is analysed by AI, which then provides suggestions based on its results. It would take a person months to develop such huge amounts of data. . Deep learning, algorithms, and datasets empower AI, which in turn streamlines the process for marketers."[6]

Artificial Intelligence and Marketing Platforms

Although prior concepts and principles have been investigated to address marketing-related issues for a long time the widespread use and deployment of AI in marketing has only recently emerged.Although AI has been applied in the majority of firms in today's world, there is still a lack of high-level implementation in many companies. Various marketers have shown an interest in using AI in the near future, with almost all of them prepared to do so fully. Only 20% of marketers have deployed one or more AI technologies in their businesses in 2018.[7]

In contrast to how the significant hype about AI may lead people to believe that they need to apply it immediately, the mismatch between preparation and execution functions as a key indicator that it is not too late to implement AI in marketing. However, AI applications in marketing are rapidly emerging, with a variety of tools and services being developed for corporations to employ for their branding. According to the analysts, following years will be a watershed moment in the use of AI in marketing.

AI has improved the marketing function, and it is now employed in practically all of the major marketing roles. AI assists in the development of sales and marketing strategies that result in significant increases in corporate performance. Pricing, promotion, distribution, and product planning and development are just a few of the marketing-related tasks where AI has been applied. According to Ali Hassan, AI is mostly employed in the digital platform, advertising, and customer relationship management. It is mostly utilised in digital marketing, such as content curation, email marketing, digital advertising, site design, Chatbots, and Predictive Analysis.

Sector of e-commerce

1. The Indian e-commerce sector is predicted to reach $ 200 billion by 2026, owing to rising internet and smartphone adoption. The online purchasing experience has been tailored for each client using AI and Machine Learning algorithms. Based on previous searches and orders, they forecast consumer behaviour and propose products that are relevant to the user.[8]
    
2. E-commerce companies utilise Al and ML to supply warehouses based on preferences in a certain geographic area. Machine learning algorithms are used to forecast future product demand and stock shelves accordingly.

Data as Product
Artificial intelligence is becoming highly common and stealthy in marketing and advertising. As a result, we are becoming increasingly conscious of how our choices are controlled and how little freedom of choice we still have. Isn't it at this time that the horrible contradiction of our more AI-dominated world becomes apparent? Indeed, contemporary Western nations were founded on the principle of individual liberty. The "autonomy" of the person, which is regarded the cornerstone of the people's sovereignty through democratic political institutions, is referred to as "autonomy."

However, our modern civilizations appear to be entrusting AI with a large portion of individual control, to the point where individuals' autonomy is finally revoked.Because many technologies are produced by firms whose business models rely on the collecting and processing of users' personal data for the purpose of selling, they pose a threat to personal liberty. As a result, the product for sale has fundamentally changed: it is no longer a thing or service, but rather the data itself. Because everything is free, the user becomes the product, our hobbies, interests, and personalities have become legal tender.

Artificial Intelligence and its danger

The type of openness depicted above eventually becomes a huge threat and security concern, as it becomes increasingly impossible to control the distribution of private data. Individuals who use these social networking sites put themselves and those who are connected to them at risk due to the amount of private information disclosed on these platforms (for example, information about other users and user networks may be indirectly accessible), and there may be a variety of risks that they are exposed to, including stalking, cyberbullying, social spamming, scamming, and clickjacking.

Users are unaware of the default settings on these social networking sites when they utilise them in their natural state. For example, in the case of Facebook, profiles are public by default until the user chooses to make them private, and because of this misunderstanding and lack of technological knowledge, malicious users and stalkers take advantage of and use the users' private information for their own gain, sometimes blackmailing them in some way. These stalkers usually threaten their victims both psychologically and physically.

1. Privacy, Security and Trust are the most important concern that is majorly in question. There is a very thin line between proper use of social media giving and pervasive use of social media, giving rise to concerns on user's privacy and security. The needs of the users are increasing every minute and this need pushing them to get more open and vulnerable to the risks of the AI that is put in the data basis. On one hand the user wants to be have many friends with whom he can share his updates and leak all his privacy there and on the other hand he wants to be safe and as private as possible, so there is a major paradox here. Hence being gregarious requires openness and transparency, but being private constricts sharing.These social networking sites provide a platform to the users to find each other and expand the friendship network.
    
2. In the current scenario AI is taking intelligent decisions and automating human activity. It's keeping a watch on all the movements of the users. AI doesn't provide easy answers or shortcuts to replace the armies of content reviewers and moderators being put in place. It is much difficult for the AI database to interpret much beyond the obvious use of inflammatory, threatening or offensive language.[9]
    
3. 'DeepFace' is a technology used by Facebook to recognize human faces in two or more than two different images in a post. Facebook even claims that the tool has more accuracy than the humans themselves recognizing human faces � with 97% and 96% success rate, respectively. In fact, its way of putting names on the faces in any image of a Facebook post has also raised some controversies regarding the breach of the freedom to move around anonymously, it has been proposed that it is a violation of Fundamental Right.
    
4. For a very long time, companies have been collecting large amount of personal information about the consumers through these social networking sites, for marketing, advertising and other business purposes. Marketing companies can predict what all are the interests of the users, for example the television shows you watch and what brand of clothing one likes. They add all these observations to the profiles of the users for marketing purposes, creating a "privacy externality". [10]

Artificial intelligence a threat to Privacy

"The enjoyment of human rights such as freedom of expression, freedom of association, and freedom of choice has traditionally required privacy." Privacy is at jeopardy these days since no one understands how social networking site data is processed or how secure it is. Artificial Intelligence gives us the power to collect, analyse, and categorise a significant quantity of data acquired by a corporation, enhancing the information collecting strategies of social actors that employ AI technology. It is critical to raise awareness about the dangers of privacy breaches.

The key reason AI is being employed on such a broad scale is because of its efficiency and speed, which human data analysts lack, and because it may be used arbitrarily.

Some significant organisations are gathering huge amounts of data in this linked world � this data is a mirror picture of a person's day-to-day existence in terms of activities, interactions, and interests. With the risk of unwanted access to one's online activity data, one's right to privacy is jeopardised. Because he lives in a smart city, even a disconnected user is at risk.[11]

Data Exploitation
To make this even worse, people are unaware of how much data their software is generating, processing, storing and sharing. As we become more and more dependent on AI and technology, the potential for exploitation will increase.

Identification and Tracking
AI may be used to recognise and track a person's behaviours across one or more devices, whether at work, at home, or in a public place. Even if the user has kept the data anonymous, the AI can de-anonymize it using inferences from other devices. This brings the distinction between personal and non-personal data to a close, which is required as part of the right to privacy.

Voice and Facial Recognition
Both two technologies demonstrate AI's brilliance; nonetheless, these strategies have the potential to jeopardise privacy and anonymity in the public sphere. The voice is recorded as data, and who knows where or for what reasons it will be utilised; similarly, the basic features of a face are saved in a database, and they may be readily misappropriated.

Predictions
"AI can infer or forecast sensitive information from non-sensitive data using powerful machine learning techniques." The AI can readily analyse their emotional state, such as anxiousness, melancholy, and worry, based on their typing pattern. Even more alarming is the fact that a person's political preferences, ethnic identification, sexual orientation, and overall health may be identified using activity logs, location data, and other such behaviours.

Profiling
AI doesn't just collect data; it also utilises it as input to sort, score, evaluate, and rate individuals according to its own method and criteria. This is usually done without the users' agreement, and the users are unaware of it and unable to fight it since they lack the means to question the rankings' decision.[12]

Legal implications of Artificial Intelligence

India is a third-world country that is acquiring access to Artificial Intelligence, and without suitable rules and conventions, it might devolve into anarchy, resulting in violations of a variety of rights and responsibilities. The Indian government intends to create a distinct set of regulations to oversee artificial intelligence, and it intends to establish a department.

However, the government may be overly reactive and overreaching at times. A recent amendment bill to the IT Act tries to restrict and enhance the government's control over the internet, which has been criticised by numerous groups as undemocratic. The need of the hour is for new regulations governing AI and its usage, as this is a highly strong technology that may be disruptive in nature and easily abused by the wrong people.

A new strategy to developing AI-related regulations might be developed by forming a new IT-AI unit that would function as a watch dog, monitoring all online activity and data transfers through various portals. Sharing AI algorithms with this cell should be a top priority in determining the nature of AI and whether or not it is damaging.

"There also is a need for Algorithm-patents to be granted to such corporations such that competitors may not steal such algorithm with illicit practices within the IT-AI cell. Such endeavours are also likely to create jobs in Indian market, especially for the infamously unemployed engineers."

Everything has its pros and cons and AI has the same case here, as it is very beneficial and on the other hand it is very much open to odds. In law, it may replace lawyers in research and can work multiple times and even faster that a human but still human supervision is required to do the same. With the advancement of AI it is very much required to have a new law and a branch of law that is AI Law, it is very likely to have this soon. Now law has more work, it has to look at the humans as well as the machines and solve the new kinds of problems arising out of it.[13]

The government is accelerating the development of rules, guidelines, and regulations controlling and regulating AI, BD, and ML.

Report of the National Institution for Transforming India (NITI Aayog)

The report recommends:

1. Creating an appealing IP regime for Al innovation, as well as the formation of a task force comprised of representatives from the Ministry of Corporate Affairs and the Department of Industrial Policy and Promotion (DIPP) to examine and issue appropriate changes to intellectual property laws.
2. Establishing a data privacy legal network to defend human rights and privacy, as well as sectoral regulatory norms in the areas of privacy, security, and ethics.

For the development of a policy framework on Al, Meity formed four committees

The committees' proposals include:

1. Creating an Open National Artificial Intelligence Resource Platform (NAIRP) to serve as a single centre for knowledge integration and dissemination in AI and machine learning.
2. Stakeholders must decide if Al systems should be recognised as a legal entity and whether an insurance plan or compensation fund should be established to pay for losses in the case of a civil liability claim.
3. Sharing of best practises - the government's use of procurement contracts to stress best practises in security, privacy, and other areas.
4. A committee of stakeholders will be formed to look at all of the issues holistically. Review of current legislation to determine what changes are necessary for the implementation of Al applications.
5. A framework should outline general principles, and companies should be free to construct internal programmes in accordance with those principles while maintaining the flexibility to adapt to changing technologies.
6. Standards for the Al development cycle must be established. The Bureau of Indian Standards (BIS) has established a new standards committee in Al.
7. The Government has proposed the development of rigorous safety parameters and setting up of safe applications are designed in such a way that it does not harm the people and property during its interaction

NITI Aayog proposed the following in the AIRAWAT approach paper:

1. Establishing a specialised Al-computing infrastructure to support the computing needs of Centers of Research Excellence, International Centers, Transformational Al and Innovation Hubs, start-ups, researchers, students, and government organisations, among others.
2. Establishing an inter-ministerial task group with cross-sectoral participation to spearhead AIRAWAT implementation.
3. The task group will look for financing to put AIRAWAT into action.

Global Efforts
The 70th session of the UN General Assembly focused on "Rising to the Challenges of International Security and the Emergence of Artificial Intelligence", and in October 2017 led a joint meeting of the UN Economic and Social Council (ECOSOC) and the Second Committee to consider the role and impact of AI on sustainable development.

The UN Interregional Crime and Justice Research Institute (UNICRI) launched its program on AI and Robotics in 2015 and declared to open a Centre on Artificial Intelligence and Robotics in the Netherlands to monitor developments in AI and robotics, with the support of the Government of the Netherlands and the City of Hague. The agency will focus and closely monitor international activity, development and uses of AI.

Earlier, the European Union published a set of guidelines for companies and governments to develop ethical applications of artificial intelligence.

The focus of UN and EU on harmful impacts of AI raises concerns for the Indian Government as the AI wave hits the nation to see into and make new laws, or at least norms and regulations. AI hits the universe at large and its necessarily for all the nations to look at it collectively and make a uniform set of regulations that the online giants follow. [14]

Conclusion
After considering all of the characteristics of Artificial Intelligence and its relationship to social media platforms, it is reasonable to conclude that AI is currently one of the most essential players in a person's life. Today's technology is completely reliant on artificial intelligence, which has an impact on social networking sites. For most Indian consumers, this is a relatively new field, and they are unaware of the ramifications. AI may be as dangerous as it is beneficial; it can disrupt a person's everyday existence and peace of mind by invading in some manner.

Unfortunately, numerous social actors, from people to companies to government agencies, can utilise these technologies against us. Our loss of privacy is only one illustration of how artificial intelligence (AI) might act against us. However, if we can correctly comprehend these technologies and their influence on our everyday lives, we will be able to defend ourselves against those who would use them for malevolent purposes.

When it comes to the rules that regulate AI and its usage by Social Media Platforms, they are woefully inadequate and lacking. There is no explicit legislation that addresses this issue; nevertheless, it is currently being developed and will be codified in a distinct law or act in the near future. There are also various interactions with other countries to look at their laws in this area so that we may embrace all parts of AI.

With regard to the protections against the exploitation of AI and the privacy of users at large, it is critical to exercise caution while uploading and sharing data online. Examine what you communicate and who has access to it; keep papers and data as private as possible; we live in a smart city, and while we can't avoid all threats, we can still avoid serious harm. Both AI and social media are in our daily lives these days and are inescapable. The only thing we can do is be alert and preventative in nature, as well as careful enough to avoid any scams or harm that may occur as a result of our use of these online portals.

End-Notes:

1. Rouse, M., Burns, E., Laskowski, N., Rouse, M., & Rouse, M. (n.d.). What is AI (artificial intelligence)? - Definition from WhatIs.com. Retrieved October 9, 2019, from https://searchenterpriseai.techtarget.com/definition/AI-Artificial-Intelligence.
2. The Impact of Artificial Intelligence on Social Media. (2019, October 13). Retrieved October 13, 2019, from https://www.manipalprolearn.com/blog/impact-artificial-intelligence-social-media.
3. Hou, Z., Teague, L., &Opitz, L. (n.d.). Social Media Powered by AI Is Here. Retrieved October 13, 2019, from https://www.convinceandconvert.com/baer-facts/social-media-powered-by-ai-is-here/.
4. Rohde, K., Vukovic, R., Zeldich, M., Ramesh, S., Hershkowitz, J., &Farkas, G. (n.d.).Benefits & Risks of Artificial Intelligence.Retrieved October 9, 2019, from https://futureoflife.org/background/benefits-risks-of-artificial-intelligence/?cn-reloaded=1.
5. Solutions, 9series. (2019, June 7). The Impact Of Artificial Intelligence On Social Media. Retrieved October 13, 2019, from https://yourstory.com/mystory/the-impact-of-artificial-intelligence-on-social-me.
6. https://www.martechadvisor.com/articles/machine-learning-ai/the-impact-of-artificial-intelligence-on-social-media/
7. Bughin, J., Hazan, E., Manyika, J., & Woetzel, J. (2017). Artificial Intelligence: The Next Digital Frontier. McKinsey Global Institute
8. Indian Brand Equity Foundation, " Indian Ecommerce Industry Report" January 2020
9. Doffman, Z. (2019, February 5). How AI Is Now Being Trained To 'Detoxify' Social Media. Retrieved from https://www.forbes.com/sites/zakdoffman/2019/02/01/googles-mission-impossible-use-ai-to-detoxify-the-internet/#68bdc84c4e5a.
10. MacCarthy, M. (2019, June 13). How to address new privacy issues raised by artificial intelligence and machine learning.Retrieved October 13, 2019, from https://www.brookings.edu/blog/techtank/2019/04/01/how-to-address-new-privacy-issues-raised-by-artificial-intelligence-and-machine-learning/.
11. Drishti IAS. (2019, May 15). Negative Impacts of Artificial Intelligence.Retrieved October 13, 2019, from https://www.drishtiias.com/daily-updates/daily-news-editorials/negative-impacts-of-artificial-intelligence.
12. Deane, M. (2018, September 7). AI and the Future of Privacy.Retrieved October 13, 2019, from https://towardsdatascience.com/ai-and-the-future-of-privacy-3d5f6552a7c4.
13. https://www.khuranaandkhurana.com/2019/04/15/development-in-technology-and-its-impact-on-law/.
14. Lapinski, K., Smith, S., Clarke, A., Johnston, B., & Nair, S. (n.d.). The Need to Regulate Artificial Intelligence.Retrieved from https://www.bbntimes.com/en/companies/the-need-to-regulate-artificial-intelligence.