Blog

While examining the history of comic books, The Adventures of Obadiah Oldbuck, written and illustrated by a Swiss professor Rodolphe Topffer (1942) is considered to be the first comic book. However, the first comic book that created buzz was Hogan’s Alley which described the life of a yellow kid on the streets of New York. It was published in 1895.

Phantom came in 1936 and Superman in 1938. These gave a Superman in 1938. These gave a historical turn to the comic books. These superheroes became popular all over the world.

In the 20th century, many comic strips reached the audiences through newspapers and magazines. Amar Chitra Katha (ACK) by Anant Pai came on the scene in 1960 and took the medium to every household. Chacha Choudhry of Pran (1971) took the trend to climax. There was an audience of children and youngsters for comic books in India.

These are the days of digital media. The entire communication happens to be through images, videos, audios and text. Visual messages influence us deeply. We have started thinking in terms of images. These images come in sequence. Sequential art is the forerunner of graphic novels.

Graphic novels have their own grammar. There are panels in graphic novels. Sound is recorded in big font. Manga comics are read in the reverse direction. The superheroes have been replaced by graphic novels.

In graphic novels, we pay attention to both the script and visuals. There is an audience for graphic novels. Orijit Sen’s River of Stories (1994) is considered the first graphic novel in India. It relates to the construction of Narmada dam. The graphic novel that created buzz is Corridor by Sarnath Bannerjee. There were many attempts to make this genre popular here in India, but these attempts were not successful.

We had a comic culture till 1990s, but later the satellite TV and OTT channels left the comic culture behind. There were serials on OTT channels based on graphic novels. That attracted youngsters to graphic novels.

We see the Manga comics re-arrival on roadside stalls. There are pictures books in regional languages. These are paving the way to the emergence of graphic novels. Graphic novels tell the stories more through the pictures and less through the words.

A Contract with God describing slum life in New York (1978) by Will Eisner is considered to be the first graphic novel. Will Eisener never claimed to be a graphic novelist. He used to consider himself as a comic book artist or cartoonist. However, when he wrote A Contract with God, he made it clear to the publisher that he is writing a graphic novel.

Comics prospered after the Second World War. They were widely available. They told the stories of Spiderman, Superman, Batman. These comics had limitations. Graphic novels rose above these limitations. Many new topics and scientific were touched in graphic novels. There was a dozen of such graphic novelists. Frank Miller (Three Hundred, Sin City), Alan Moore (From Hell, Watchman, V for Vendetta), Marjane Satrapi, (Persepolis), Jonathan Entiwistle (The End of the Fucking World) presented the serials based on graphic novels. This made graphic novels popular among the common public. These books acquired a literary dignity.

The Govt proposes to fund 50 per cent of the cost of creating AI compute infrastructure in the country under India AI Mission (funds allocated Rs.10372 crore). The compute capacity will be created as public-private partnership (PPP). There will be GPU based servers for startups and innovation centers.

The target is to establish at least 10,000 GPUs worth of AI compute capacity in the country. GPUs are essential for creating AI models as these require large scale computing, which cannot be done by CPUs.

To expedite the process, the government is also looking at the viability gap funding.

The government for its own use is planning an independent capacity through the National Supercomputing Mission.

We have a large number of STEM-trained graduates. Many of them have exposure to AI. We thus have the right type of manpower to use AI for real life use cases.

Generative AI is a useful tool in drug development. India so far focused on generics and low-cost vaccines. It has yet to be competitive in mainstream pharma market. India can scale up using AI.

It is difficult to compete with the US and European drug majors who have abundance of resources. India struggles to progress a drug beyond phase II clinical trials.

China used AI in new drug discovery to move ahead. China made a goal to reach human clinical trials by leveraging AI. In preclinical stages, a molecule was identified, and novel drug candidates were assessed for their binding efficacy with the molecule target. The preclinical trial outcomes were predicted using AI as a tool.

MIT discovered a new antibiotic that could be used in drug-resistant infections by using an ML algorithm. They bypassed traditional experimental approaches by rapidly screening a million plus chemical compounds.

Merck and Pfizer too leverage generative AI. Merck’s proprietary platform is ADDISON which is used for drug discovery. Pfizer uses AI-powered chatbots to deliver personalized messages to clinical trial participants.

AlphaFold (Google DeepMind product) predicts protein structures. Amino acid chains of proteins fold up into 3-D shapes. This determines its functions. A protein’s structure can alter its behaviour by introducing a drug that binds to protein. Traditional methods to map protein structures have limitations. So far only 20 per cent of human body’s 20,000 proteins have been mapped (2022). AlphaFold will solve this problem. It can predict the points where folds would be optimum.

Nvidia, the chip maker, has made BioNeMo — a generative AI platform. It facilitates for pharma companies the development of foundational models for drug discovery. It reduces unnecessary experimentation, and, in some case, this gets completely replaced.

Nvidia has partnered with Recursion, Amgen and Roche.

Gen AI is revolutionizing health sciences sector in India. Gen AI , it has been projected, could contribute $4-5 billion addition to GVA of the Indian pharma sector by 2030.

In fine turning, we deal with pre-trained model, say an LLM or image classifier. It can be trained further on a specific task or dataset. The model then adapts to a new task by adjusting its parameters. At the same time, it leverages the knowledge it gained while being pre-trained. Thus, there is transfer learning. Transfer learning is an ML technique where a trained model on one task is repurposed or adapted for a different, but related task. There is no training from scratch. Transfer learning allows the model to use knowledge learned model to use knowledge learned from one task to another task, typically with less data and computation.

To illustrate, an image classification model trained on a large data set (with millions of labelled images) can be adapted to distinguish between various species of flowers. However, there is only a small dataset of labelled flower images. One can use transfer learning. A pre-trained model is fine-tuned on smaller dataset of flower images. The model transfers the knowledge it has gained previously of recognizing general features in images such as edges, textures and shapes. It has learned this from a large dataset, and it can quickly adapt to the specific task of classifying the flowers, though the data is limited.

TikTok is the Chinese social media app which in less than a decade has risen to global dominance. It is a short-form-video-sharing app. India banned it in June 2020, when the armies of China and India came face to face. The USA took similar steps — ByteDance, the parent Chinese company was asked to divest from the company or be booted out of the US.

Before the ban India was the biggest market for TikTok. The US now is the biggest market — 143.7 million monthly users (January 2024).

While banning, India cited the data security and safeguarding privacy as the reason. The US also feels that it gives access to the data of Americans. The European union banned TikTok on the devices of its staff. Australia and Taiwan has taken similar steps. Pak bans it multiple times for indecent content. Nepal too bans it.

Though wholly or partly, TikTok is banned, it has not dented its relevance or popularity. Social media platforms have launched TikTok clones — Reels and Shorts from Facebook and YouTube respectively. Even LinkedIn is testing a short-form video platform. India has indigenous versions of TikTok — Josh, Chingari and Moj.

Social media platforms have also adopted vertical video feature, with scrollable feed — for instance the Discovery feed. Instagram YouTube and Netflix too experiment with the vertically shot scrollable short-form video feature. Thus, it is TikTokisation of the social media.

What matters is that TikTok resonates with the users. Though TikTok is a late entrant, it could garner 1.5 billion monthly active users (2023 end). It remains dominant in both tech and non-tech space.

AI has created a lot of buzz since it has the potential to affect business and our behavior. AI has been a part and parcel of marketing and advertising for a while. To formulate business strategies, AI platforms and tools have been used on data. There are predictive algorithms for business strategies and campaign optimization. The advent of LLMs and generative AI has led to applications in consumer insights, strategy and creative.

There are two dimensions to AI in business. One is the way we work, and second is the work that is created. AI brings efficiency in the business process with its tools. In the work, AI tools are used to imagine and create magical ideas.

Agencies will continue to invest in technology. A layer of AI will be infused across the organisation. AI will make people more efficient and productive. Many things that we will be doing tomorrow are not being done today. There is no conflict between technology and creativity

In advertising, one requires soft skills of understanding human emotions and needs. Advertising can bring about behavioral shifts. This is done through creativity. Technology is here an enabler. It brings creative ideas to life. It tests these ideas. It iterates faster. One needs both — technology as well as creativity.

It is very embarrassing for talented people who are successful to admit their mistake and then seek assistance to fix it. Tech giant Nvidia CEO and cofounder Jensen Huang did both this to save his company.

At present, Nvidia is valued at $2.2 trillion, and attributes its success partly to AI boom and huge demand for its GPUs. Nvidia was founded in 1993 and just three years old in 1996. It was in financial distress since its contract with a gaming company Sega had fallen apart.

Huang wanted Nvidia to remain a going concern. Under the contract, he was making chips for the gaming company Sega. Sega was a major client that funded Nvidia. Nvidia made low-cost chips deviating from the industry standards, After being into it for a year, it realized that its architecture was sub-standard. To add fuel to the fire, Microsoft rolled out DirectX software interface — it then became an industry standard for gaming platforms. However, this was not compatible with Nvidia chips.

On completing Sega’s game console, they would have introduced inferior technology incompatible with Windows. Nvidia would have become a laggard who would have been far behind to catch up. And if they violate the contract with Sega, they would be out of business. Either way, they are the losers.

It was necessary to explain the situation to Sega and make a clean breast of it. It was necessary to suggest to them to find a new partner. Despite severance, Sega should pay Nvidia in full or else they would be out of business. Huang’s situation was embarrassing, but he faced it. Sega agreed, and that gave Nvidia some breathing space.

Sega’s funds were used by Nvidia and it developed a new chip — the RIVA 128 that was compatible with Direct X. Sega used alternative chips from Imagine Technologies — Power VR for its Dreamcast consoles.

Nvidia successfully sold 1 million new chips as they had higher graphic resolutions (1997). It turned around Nvidia’s fortunes.

Admitting a mistake and asking client for empathy is so difficult. These are the hardest traits, especially for the talented and bright guys. Swallowing one’s pride was the right thing to do for Huang, and he did it.

As soon as ChatGPT was launched on November 30, 2022, there was an avalanche of new generative AI models. Some of these were chatbots powered by LLMs, a variety of plugins, APIs, virtual assistants and copilot aids.

In 2023, we came across multimodal foundational AI models. The trend continued in 2024 — text-to-video sensation Sora (mid- February 2024 launch, Alibaba.s EMO).

The models have become sophisticated — greater parameter count, larger context window, greater computation efficiency. Gemini Ultra (Google) and Llama 2 wanted to outperform GPT-4. Anthropic and Mistral too wanted to do the same thing.

There was a quick replacement of models — Anthropic’s Claude 2.1 followed Claude 1.3 in four months. Facebook’s Llama 2 came three months after Llama 1.

AI models grow in size. There is huge data requirement. There is an increasing demand for data centers. It raises the cost of building advanced foundational models (FMs).

Open AI invested $4.6 billion to develop GPT-3 (2020). GPT-5, the proposed model could require an investment of $1.25-2.5 billion just to train the model. The cost of product failure could be amazingly high.

OpenAI has acquired a large customer base, and it retains the base. It has enlarged the base substantially. It has used easy to use plugins, virtual assistant apps and image generation models. ChatGPT has adopted for-profit strategy by introducing subscriptions. OpenAI is thus attractive for investors and for valuation experts.

Gemini’s setback has alerted others. Governments may ask organizations to take prior permission before the release of the model. The governments may expect certain disclosures and call them under testing if they are being evaluated.

A full-fledged use will be contemplated after the model is fully ready. There could be some changes in the training regimes. There could be increased scrutiny of unsupervised learning. Even supervised learning may have to comply with standards and protocols.

The UCPMP, 2024 follows the code previously released in 2015. The previous code was ineffective to curb the unfair practices by the pharma companies while dealing with doctors. The monetary incentives and freebies to the doctors add to the cost of the drugs which ultimately burdens the pockets of the patients.

The present code lays guidelines for the conduct of medical representatives (MRs) who promote products by distributing literature, brand reminders and free samples. They establish a relationship with healthcare professionals.

There is a provision for lodging complaints against unethical marketing practices. The responsibilities of the chief executives of pharma companies have been defined in such matters. The code extends to medical device makers.

However, the code lacks the teeth as it is not punitive and is purely voluntary.

The code prohibits gifts, sponsorships or payments by companies (or their distributors, wholesalers and retailers) to healthcare professionals. Doctors can associate with companies as consultants-advisors for research services only. Pharma firms cannot fund travel and lodging expenses. When doctors attend conferences on invitation, all the details must be uploaded including the funding-expenditure costs on the website. This will be subjected to special audit.

Pharma associations will establish an Ethics Committee for Pharma marketing Practices. (ECPMP). If the code is violated by an entity the committee can take action for the first time. The appeal can be raised before the Apex Committee for Pharma Marketing Practices (ACPMP). It will be headed by the Secy, department of pharmaceuticals and will also have a joint secy and a finance officer as its members. The ACPMP’s decisions will be final and binding on both the parties. It is empowered to impose penalties or make reference to an appropriate government agency or authority.

It is to be seen whether a self-regulatory mechanism is enough.

At the end of 2023, TV ad revenues were 26 per cent of the total ad revenues of Rs.2.32 trillion. It is a fall of 36 per cent since 2019. In fact, 2023 was a bad year for TV, but 2024 could be worse, as digital media would overtake TV. This is inevitable since content is being increasingly consumed on smartphones.

However, there is still a loyal to TV audience of higher age group that is accustomed to appointment-based viewing and consumes TV content. Besides, TV’s larger screen is an altogether a different experience. It makes TV relevant. Advertisers who want to reach mass market of traditional lean-back audience find TV convenient.

TV has unprecedented reach and ease of use. It is thus very good at building brands — especially brands at awareness stage which require continuous reinforcement. Brands being promoted on TV have strong recall value.

TV also is good for sports marketing. Of the total sports audience of 678 million, a high 51 per cent still prefers to watch sports on TV. For some time, sports will co-exist both on TV and OTT. TV provides an experience of community viewing, whereas other screens are viewed solo.

Slowly, TV homes will upgrade to pay TV homes.

There is action on internet-based and broad band video market. There is a demand for streaming services (Netflix, Prime Video, Hotstar). This has disrupted TV. YouTube provides video on demand — it has 61 per cent of TV’s reach. In some areas, such as Delhi, UP, Bihar and Northeast, it has reach similar to TV or even higher.

Nonetheless, TV will continue to be an important channel for advertisers. TV screen will grow to 202 million by 2026. Many connected TV screens will be added here. There will be migration from linear to connected TV. It will add to viewership. It will attract more advertising. Video on demand will be seen both on mobile devices and connected TVs.