Category: Uncategorised

The spread of knowledge gained speed with the invention of printing machine. Typewriter made writing professional and computer made it personalized. Each leap triggered anxiety first and later empowered us. However, AI with tools such as ChatGPT raised the issue of whether what we are outsourcing is just what we have to write or the process of thinking itself. As AI gains traction in education system, there is research on the hidden cognitive costs of AI-generated writing.

MIT experiment asked the students to write an essay unaided, using a search engine and using a Chatbot. The students wore electro-encephalogram (EEG) to record the activity of the brain. The students who took ChatGPT assistance manifested the least activity of the brain. Their essays were generic and soul-less. They could not own their work later. Unaided writers showed the highest cognitive engagement and had a sense of authorship. This shows that cognitive ability is affected depending on the use of AI to think for us. Apple researchers conducted another study. They endorsed MIT findings. ChatGPT writing mimics human writing and uses statistical pattern-matching. It fails to reason in the face of unfamiliar and complex problems. As LeCun from Facebook puts it, ‘AI does not understand the world, nor does it truly think.’

Additional research elsewhere too confirm these findings. AI tools bring in a decline in creativity, loss of personal voice, and dilute writing confidence. It is delegation of mental work to AI — cognitive offloading. We fail to build the muscles of analysis, synthesis and articulation that education is all about.

Should we ban AI in the school system? Not at all. The issue is how it is used. It could be used for grammar correction, summarization or refinement of writing. It should not replace the original thought. Students should be encouraged to write unaided first. Later they can use AI for editing. The ability to think, research and marshal the ideas should be retained. The NEP also champions critical thinking and creativity. AI must facilitate these and should not eclipse these.

Tesla has launched robotaxi service this week (June end, 2025). Handful of Model Y robotaxis are offering paid rides in Austin Texas. It is a milestone. Tesla claims that the cars it sells have already the hardware needed to be robotaxis, and its self-driving software trained on a vast fleet of existing cars driven by beta tester unpaid drivers who can handle virtually any situation on the road. Yet Austin launch has been defined by its boundaries (not by its potential) operating in a limited area with invite-only customers. Besides, there is safety-monitor in each vehicle.

Tesla’s core EV business profits are on the decline. Robotaxis and self-driving technology can pay off 10-15 years from now and the business does not make money today.

There could be a good market for robotaxis at some point in future. The competitor Waymo has adopted city-by-city approach.

Regulators are looking into traffic violations captured on video, but these are teething troubles.

The issue is how quickly Tesla can commercialize this. It has the edge — it uses sensor-lite approach, avoiding LiDAR. That makes robotaxis economical. However, it should scale up soon. It should reach 2.5 lac rides a week. The higher vehicle costs can be offset by more paid miles. There could be other competitors — Zoox from Amazon.

Maybe, one year from now, robotaxis may be rolled out in multiple cities. Yet today, the launch does not inspire confidence. Tesla also cannot defer the dream.

There are AI talent wars — a no-holds-barred fight for the best people. Here the employees hold all the cards. Top engineers are being courted. Facebook is going after talents with ‘pay packages of up to $300 million over four years.’ Irrespective of the size of the packages, the aggression is causing considerable anguish among company leaders. It is like feeling someone has broken into our home and stolen our assets. The challenge is not limited to large companies. It is possible that those who are being hired may not build a product, but may wait to get the term sheets to start their own company.

OpenAI pitches for AGI to prevent their talent to drift away. Employers must convince the employees that their work will make it to history pages. Whether employees will buy this pitch is a different matter.

India wants to promote manufacturing in order to create a vibrant economy and generate employment. The government and industry are trying to make manufacturing shop floor manpower all-inclusive. India has set aside one of the largest gender budgets — 8 to 9 per cent of the total union budget in 2025. There are dedicated schemes such as Sakhi Niwas (under the Ministry of Women and Child Development) for working women’s hostels. Worker housing with gender sensitive design is also being promoted (under the Ministry of Housing and Urban Affair).

Indian’s formal economy remains male-dominated. As of May 2025, India’s Worker Population Ratio for women in urban areas remains just 23 percent. Behind this modest figure, there are lot of changes in the working conditions. Women in Dr. Reddy’s experienced headaches, and they were provided lighter helmets made specially for women. At CEAT plant, women now work in night shifts by restructuring factory infrastructure. Women are picked up and dropped, by company vehicles, accompanied by female security guards. Work-stations have been designed ergonomically. There are onsite creches. There are flexible shift options. There is women empowerment forum. All this has resulted in better productivity.

Tata Motors, Pune plant has an all-women assembly line. Work-space has been redesigned for women. There are restrooms and creche facilities. There is last-mile transportation system. They prepare women for leadership roles. In a cafe, employees and managers engage in open dialogue. They resolve work-place issues.

Women’s labour force was 30 per cent in 2019-20. It has increased to 41.27 per cent in 2023-24. Still, urban participation lags behind — 25.3 per cent in May 2025 (as against 35.2 per cent in rural areas).

There are still many companies in MSME segment where gender-inclusive workplaces are an exception. There are issues of infrastructure and social norms. Women are reluctant to work on shop floors. But this is changing very fast. There should be basic amenities such as clean toilets, last mile transport, safe accommodation and creche facilities. Large companies can afford to invest in these changes. MSME consider women as costly hires.

Electronics manufacture employs 80-100 per cent women on the shop floor.

Venture capitalists invest in startups to benefit from the returns of the sale in future. Big Tech, these days, do not buy startups but do aquihiring the promising AI teams. They are after the most valuable manpower talent and would like to avoid anti-trust scrutiny by leaving behind business operations. It is a way to eliminate competition economically.

Microsoft acquired Inflection AI’s 70-strong team in 2024. Meta bought a 49 per cent stake in Scale AI for $14.3 billion and employed its CEO Wang to head Superintelligence Labs division. Google paid $2.4 billion for the senior leadership team and licensing rights of Windsurf, an AI coding assistant.

Acquihiring is not a new phenomenon. Over a decade big tech firms paid a few million dollars to hire talented engineers and product teams from startups leaving their investors with modest or no returns. The strategy was implemented more after the race for generative AI began.

The M&A teams at tech giants are exploring all types of hybrid deals. They want the deals to work, as anti-trust regulation works against them. It is to be seen whether in future acquihiring makes markets healthier.

AI companies need cloud platforms for several reasons related to scale, performance, flexibility and cost-efficiency.

First of all, all large AI models such as GPT or image-recognition systems require powerful chips (GPUs or TPUs) and computing infrastructure for model training. Cloud providers offer on-demand access to this power. There is no need to invest and maintain costly hardware. Such cloud providers are AWS, Google Cloud and Azure. Besides, cloud environments make it easier to train models across multiple machines or nodes. This accelerates the process.

Secondly, AI models need huge amount of data. Cloud storage allows data storage that is secure, distributed and scalable. It also supports real-time data ingestion and processing.

Thirdly, based on demand, the company’s infrastructure can be scaled up or down. This elasticity helps startups as well as large enterprises.

Fourthly, pay-as-you-go pricing avoids upfront infrastructure investments. It results into cost efficiency,

Since cloud provides security, encryption and compliance with regulations, it is useful while handing data in healthcare or finance.

Clouds have data centers worldwide and the models can be deployed close to users.

Lastly, clouds offer ready-made AI-ML tools.

Microsoft has strategic partnership with OpenAI and provides OpenAI exclusive access to Azure’s supercomputing infrastructure. Microsoft has invested heavily in AI supercomputing infrastructure. It has Nvidia GPUs — A100s or H100s. It has high-speed networking and distributed storage. It has specialised hardware for training LLMs. OpenAI’s models run via APIs running on Azure infrastructure.

Google’s DeepMind uses its own cloud and AI infrastructure, benefiting from TPUs, Borg and Kubernetes systems. It has data storage and inter-connect.

AlphaGO, AlphaFold, Gato and Gemini use Google’s distributed computing and data centres.

A Chinese car maker Xiaomi has launched a luxury sport utility vehicle, the YU7. It has stirred the market by indicating what future holds for the Western automakers. YU7 is a stylish tech-laden SUV with a driving range of 835 km. Its affordable price starts from $35000 for its entry-level version. YU7 competes with Tesla’s Model Y in China. It is not yet available in the US or Europe. It poses a risk to Western car makers with higher sticker prices for their luxury cars. China is making sophisticated EVs now. Investors feel that Ferrari should hold its fort, while Xiaomi continues to make a cheap Ferrari EV. Germany’s Porche and other car makers should be more resilient.

Luxury cars could become commoditized. First of all, initial pickups have become commonplace in EVs. There is faster innovation in EV space. The product development cycles are cut short. Consumer perceptions are changing. Chinese manufacturers have introduced voice recognition, advanced software and AI in their vehicles. Customers are not ready to pay a premium for such features.

Western automakers products are not price competitive. Mercedes $160,000 electric version SUV is costly. It is heavy, has limited range and towing capacity. There are many EV flops. Luxury EVs such as Porsche depreciate too fast. Some Western automakers do not see the models from China to be competitors in next five years and are relaxed about their achievements on the racetrack. They are ready to cut prices. The US is off limit to China at present. Chinese brands have made limited inroads in Europe. The way the European luxury car makers can depend on their costly models in the era of electric software-defined models is by developing models that surpass the Chinese models. And it is a tall order.

Basically, derivatives are risk sharing tools. These are used by investors, businesses and institutions. Each derivative is valued on the basis of its underlying asset or variable. These assets or variables could be stock, commodity or even the temperature at a place. Farmers can hedge the decline in the prices of wheat. Airlines can manage aviation fuel costs. Exporters seek protection against currency swings. Strictly speaking, these are financial instruments, and yet they serve an economic purpose. They transfer risk from those who bear it to others who are willing to take it.

There are well-designed weather-derivative contracts. These reduce uncertainty and support long-term investments.

Weather becomes tradeable when it is location-specific, measurable, time-bound and independently verifiable. It should be reduced to a number agreed upon by both the parties before trade. At the time of settlement both the parties accent it without dispute. The most commonly used weather variables are :

Temperature: Here Heating Degrees days (HDD) and Cooling Degree Days (CDD) are measured indices. They track how much daily weather deviates from a set base.

Rainfall: The contracts are structured on cumulative rainfall, say total mm over a month in a district, or deviations from long-term averages.

Snowfall or Wind Speed: It is common in advanced markets with exposure to ski tourism construction or wind energy projects.

Trading of Weather Phenomena

These are not traded as weather events but are traded as numerical outcomes. To illustrate, a rainfall futures contract might pay Rs.7000 per mm shortfall if the actual rainfall in Sawan drops-below the 100 mm threshold in Satara. The data source is agreed upon beforehand.

There are informal markets which operate on weather cues. These are instinctive. Weather derivatives structure this.

Using Weather Derivatives

A cold drink company expects a summer surge in sales, but if summer experiences unusually cool weather on account of rain spells undue, there could be a drop in demand. Insurance does not cover this. Here a temperature derivative could help. If May temperature stays below 36-degree centigrade, it creates a cushion to offset lost sales.

Farmers can take rainfall-linked hedge, say a rainfall 30 per cent below average. The derivative then pays. Banks can use rainfall-based hedges against its rain portfolio of loans. Lending then becomes viable.

Well-functioning weather derivatives market enhances credit quality for financial institutions. There is risk resilience.

Insurance vs. Derivative

When the loss is catastrophic, insurance is more helpful than a derivative, say damages to a warehouse on account of storm or floods destroying inventory. Derivatives are suited for non-catastrophic, recurring risks, say low rainfall or cooler temperatures. These affect revenues but not assets. Insurance safeguards property. Derivatives shield cash flows.

Corporates raise money from capital markets and use that money to buy cryptos. Their share prices raise, and that makes them repeat the whole exercise. MicroStrategy, a software company, owns Bitcoin worth $64.8 billion. Companies vie with each other to build their own crypto stash. It seems there is a race to build crypto corporate treasury.

A coffee company whose shares had fallen 90 per cent in 2024 have doubled in value since it went in for a crypto buy in May. It holds 69 Bitcoin, and wants to add thousands more.

Most corporate treasurers are risk averse. They would not buy volatile tokens. They too are lured by crypto wave. There is investor demand for exposure to crypto vehicles, it makes sense for more players to hop aboard the bandwagon and offer supply to match.

There are risks. A crypto sell off could turn into a crash. These are overpriced tokens held by the weak and indebted companies. It could create a wave of forced selling. It should be remembered that in the last crash, Bitcoin fell more than 50 per cent.

Are we using the Bitcoin innovation which is blockchain based prudently? Or is it speculative frenzy to make ‘money from money’?

Hard-boiled candies costing around Re. 1 or even less are popular both with the children as well as adults. These candies are widely distributed through grocery stores and other small retail outlets.

Pulse candy was launched by Dharmpal Satyapal in kachcha aam or raw mango flavour a decade ago. Within a year of its launch, it became a Rs.100 crore brand. Last year ending in March 2025, it became a Rs.750 crore brand. It is growing at the rate of 15 per cent (against the industry average of 9 per cent). DS has a share of 19 per cent in the Rs.4000-crore hard-boiled candy market. Pulse brand is likely to reach the 1000-crore mark in a couple of years from now.

The competing brand Alpen Liebe has advertising and marketing muscle, but pulse does not. It focuses more on product innovation and consumer experience. Pulse did some clever packaging. It grew despite no marketing push and has become a case study at IIMA.

It bets on bold Indian flavours. The market is replete with sweeter Western flavours. It expanded the market from kids to adults. It finds flavours that appeal to adults. Its major revenue is derived from adults. Its growth is organic. The brand is an outlier. There was raw mango flavor in the market prior to Pulse. But it was launched as a combination of sweet and salt, mimicking raw mango that most of us savoured in childhood. There is salt in the center, which refines the candy experience. It has become its best-selling candy. It introduced other flavours also such as litchi, orange, guava, and pineapple. Pulse has also launched tamarind flavor that makes Indians nostalgic about their childhood.

Pulse is an indulgence at a modest price. It has remained non-preachy. It relies on buzz word and school-yard gossip. Its signature is its masala core. It could explore more extensions, say lollipops and chews.

Its competing products are Perfetti Van Melle’s Alpen Liebe, ITC Candyman, Parle Products Toffees and Ravalgaon. Perfetti is the largest competitor which earned a revenue of Rs.3500 crore. DS brand Pulse is around Rs.1000 crore, and the company wants it to reach Rs.5000 crore by 2029.

Though initially marketing was not the focus, the marketing spend has now reached 6-8 per cent of the annual revenue, on par with industry norm. Below-the-line promotion and distribution matter more in confectionery industry. Sampling and influencers have worked in favour of Pulse. It reaches 35 lac outlets in the country.