The purpose of the trademark law, considering the analysis drawn by the Supreme Court in Dau Dayal Vs State of Uttar Pradesh AIR 1959 SC 433 is “ to protect the rights of persons who manufacture and sell goods with distinct trade marks against invasion by other persons passing off their goods fraudulently and with counterfeit trademarks as those of the manufacturers”. The trademark law serves its purpose efficiently to ensure undistorted competition and transparency among buyers and sellers, which means that in addition to protecting consumers against confusion it also ensures fair competition and proper functioning of markets. But with the radical advancement in technology, the elementary features of this law stands challenged.
Until the last decade we have seen seemingly not witnessed any major change or innovative disruption in the consumer purchase behaviour which remained almost human driven. Initially product selection by customers were influenced by shopkeepers who were well conversant with the different products in the market. With the establishment of self-service stores and shopping malls, the decisions on product selection became based on brand loyalty, brand reputation or guided by shop assistants. Hence there has been dominance in accepted concepts of trademark law when considering the comparison of a trademark from the perspective of possible misuse/unfair use such as ‘average consumer’, ‘imperfect recollection’, ‘likelihood of confusion’, ‘concepts of aural, conceptual and visual comparison’ and the ‘slurring’ of trademarks. In the famous case of K. R. Chinna Krishna Chettiar vs Sri Ambal & Co., Madras & Anr, it was held by the Supreme Court that the words ‘Ambal’ and ‘Andal’ have great phonetic and deceptive similarity and that there is a actual threat of confusion between the marks ‘Ambal’ and ‘Andal’. The Court has also considered and discussed the aspect of consumer recollection/ average buyer with imperfect recollection before arriving at such decision.
However, with digital transformation and emergence of ground breaking innovative technologies such as Blockchain, Robotic process automation (RPA), Artificial Intelligence (AI)/machine learning, Data Analytics, Internet of Things (IoT), the purchasing pattern of consumers has undergone a substantial change.
According to the research and advisory firm Gartner, Inc. the estimated global value of business derived from is expected to rise from $1.2 trillion in 2018 to $3.9 trillion in 2022.
Sébastien Szczepaniak, head of sales and e-business at Nestlé, in an article in The Wall Street Journal (written by Saabira Chaudhuri and Sharon Terlep, February 2018) predicted that within five years, as a direct result of the rising use of AI assistants in retail, 50% of searches for products will be done by voice. 
Hence with more AI guided purchase on rise, in our article we intend to study the impact of such AI guided recommendations and purchase vis-à-vis trademarks laws.
INVOLVEMENT OF ARTIFICIAL INTELLIGENCE IN PURCHASE DECISIONS AND ITS RELEVANCE TO TRADEMARK
AI operations can be commonly seen in online retailing and the simplest example of its presence is the Amazon website. The AI technology used by Amazon helps it to make product recommendations to the customer through its AI enabled assistant based on several factors like the customer’s browsing and purchase history.
Amazon echo and Google home devices are other devices which have incorporated AI, make product suggestions to the users depriving them of first making their own product selection. The AI assistants like Alexa in Amazon Echo, makes decisions for selecting and suggesting products to the users based on the analysis of the available data.
Amazon Dash provides replenishment services by ordering consumable products which are running out of stock.
Chatbots are another example of automated assistants which interact with customers enhancing their online purchase experience. One may come across Shopbot while shopping on eBay which suggests personalized products to customers and helps in customer service.
The consumers thus may come across a range of products while searching online but with AI assistants making recommendations, it will offer specific brands based on factors like user's visual preferences, past behaviour of a customer, behaviour of similar users and similar other factors.
The increase in use of AI assistants could present new practical difficulties for trademark holders. The aspect of Consumer loyalty is limited only to human beings. But can this be achieved for AI driven purchases. The Article in The Wall Street Journal, referred to above, acknowledges a fresh concern of the rankings of goods that are recommended by these AI assistants. In the abovementioned article, Szczepaniak has opined that you need to be “first position or you go home”. Saabira Chaudhuri, one of the authors of The Wall Street Journal article, recognized seven brands of toilet paper that were on offer in her local grocery store and pointed out that on Amazon’s website there were above 30,000 results for toilet paper whereas AI assistant Alexa offered simply two brands as a result of her voice search for toilet paper. Although undoubtedly the choice of brands on Alexa and AI assistants will increase, and Amazon is circumspect in identifying the exact basis of its ranking on Alexa, it has identified popularity (sales), rating, reviews, price, delivery speed and of course being available via Amazon Prime as factors in determining whether a product is available on Alexa and how it is ranked. 
Not undermining the capability of AI in making the retail experience easier, this AI technology is eliminating the humans from being an important part of the product identification and selection process and/or there is gradual replacement of human consumers with AI. So while onus of accurate search was on consumers and the concept of phonetic, visual and structural impact was a key factor while making comparison between trademarks but with the intervention of AI technology several questions arise. For instance- can AI be deceived and confused when the search is voice based? Does AI have imperfect recollection and can it be considered as an average consumer? Is it capable of distinguishing counterfeit products from genuine ones and if not is it liable for secondary infringement?
Another important question that needs to be addressed is that of liability and responsibility of AI. There is a lot of uncertainty as to whether AI is liable for its actions and if yes then to what extent or is it the creator who is to be held liable. It is also unclear whether the product/work of an AI enabled system is a result of its own intelligence or should the person creating it be held responsible for the product/work. For instance, AI is also used now to create artistic or literary work and who has copyright over such work is an unanswered question.
A study of the Google France cases (CJEU, March 23 2010) and the case of L’Oreal v. eBay International (CJEU July 12 2011) has made it clear that the AI application provider is likely to be held liable for infringing activity on their platforms in the event they are actively aware of such activities and take no preventive steps.
Similarly, in the case of Cosmetic Warriors and Lush v Amazon.co.uk and Amazon EU ( EWHC 181 (Ch)) amazon was held to be responsible for the infringement activities. 
It is therefore unclear whether Trademark law will be able to cope up with the new AI technology and we need to assess how to keep these laws updated in order that it goes hand in hand with the new revolution. What is indeed clear is that the underlying principles of Trademark law will undergo a drastic change and every concept will need to be redefined with new cases.