Futuristic technology feels like science fiction. Google has unveiled, as we have observed in a separate blog, its quantum computer chip called Willow. Google’s CEO Sunder Pichai has floated an idea of taking quantum computing to space.

Traditional computers are very slow as against quantum computers. There are several challenges in quantum computing. Quantum chips are isolated so that there is no electromagnetic interference. Besides, quantum chips must operate at extremely low temperatures. These conditions are difficult to satisfy on the earth. This hinders their scaling.

Elon Musk’s company SpaceX has introduced Starship, and he is ambitious about the lunar and Martial landings, and colonization in future.

Sundar Pichai suggests that quantum computers must be sent into space to take advantage of conducive conditions. There is no interference of the earth in space. The stable low temperature also could be maintained. Thus, if the quantum systems are in orbit they can process data and conduct research. Elon Musk has responded positively to this proposal.

The conversation was not restricted to quantum computing but was extended to solar energy. Pichai hinted at future technologies such as Dyson spheres.

The ideas are still nascent. However, they have potential benefits — climate modelling, healthcare and material science.

Entanglement is a crazy phenomenon. Apparently, it allows particles to communicate over vast distances, faster than the speed of light.

Though there is connectivity between entangled parties, these do not necessarily share information between them.

In quantum mechanics, a particle is not really a particle but a cloud of fuzzy probabilities. These probabilities describe where this particle could have been placed, when we look for it.

These fuzzy probabilities are in fact quantum states. If the certain circumstances occur, we can connect two parties in a quantum way. Therefore, a single equation describes both sets of probabilities at one and the same time. This by itself is called entanglement.

Quantum state particles show similar properties. We know electrons, the sub-atomic particles, revolve around the nucleus of the atom. Let us look at their quantum spin. The spin could be in one of the two states — either up or down. If these two electrons are entangled, their spins show correlation. The entanglement can be planned in a certain way, so that the spins are always opposites of each other.

Examine one particle randomly. There is random spin that points upwards. Can we predict about the second particle? The quantum state has been carefully arranged. Therefore, we are sure that the second particle must be pointing downwards. The quantum state was entangled with the first particle, as soon as one declaration is made. That means both the declarations are made.

What happens if the second particle is at a distance? In quantum theory, as soon as one particle is chosen, the partner particle instantly knows what spin to be in. Apparently, it is communication faster than light.

If two different persons take measurements for particles, the person measuring particle A knows the spin of particle A. Or else the other person tells you about the measurement. In both the cases, there is no transmission. Either you measure or wait for the other person’s signal.

In case of two connected particles, there is no advance knowledge of anything. The person knows how the particle is behaving but I communicate it slower than the speed of light.

Thus, entanglement process, though instantaneous, the revelation of it is not instantaneous. The older communication method prevails to piece together the correlation of the quantum entanglement.

We have already discussed teleportation in previous blog — it is transfer of information from one place to another by using what is known as quantum entanglement.

There is a pathbreaking achievement in quantum teleportation. Quantum teleportation transfers information instantly and over any distance. At Northwestern University, Illinois, US, researchers demonstrated quantum teleportation over optic fiber cable (OFC).

This is achieved through quantum entanglement — an invisible twin connection between two particles linked in such a way that what happens to one affects the other, even if the particles are miles apart. You are not sending physical objects. You are sending the state or condition of the particle.

It is so exciting. It was unthinkable. It paves the way for next generation quantum and classical networks sharing the same OFC infra. It pushes quantum communications to the next level.

The experiment chose the delicate quantum information carried by photos. It should survive the clutter of internet data. They utilized, therefore, a specific wavelength of the quantum signal. There were filters to prevent interference from other data streams.

The photons were placed at a judicial point so that there is least scattering. There was no interference from the classical channels which were present.

They successfully teleported quantum state of light. It is a big leap forward since it integrates quantum communication to existing internet work. It is not just a simulation. It happens in real-world situation.

It occurs at the speed of light. It is a step to quantum internet. It could revolutionize traditional computing. The technology has immense potential.

Weight-loss treatments have seen remarkable breakthroughs. A whole new category of drugs called glucagon-like peptide-1 (GLP-1) has emerged to reduce blood sugar level and promoting weight loss. Semaglutide (Ozempic) of Novo Nordisk was approved by the US FDA in 2017 for treatment of type 2 diabetes. It soon gained attention for off-label use in weight loss. Novo Norolisk launched Wegovy, a Semiglutide pen-injection in 2021. FDA approved it for long-term weight management.

In November 2023, Eli Lilly introduced tirzepatide (Zepbound) as a GLP-1 receptor agonist and received the FDA approval. This followed the success of tirzepatide of Lilly (Mounjaro), an injectable anti-diabetic, also used for weight loss off-label.

Zepbound and Wegovy both boost GLP-1 level in the gut and brain, making the user feel full (thereby reducing appetite). Zepbound in addition also enhances the levels of GIP (gastric inhibitory polypeptide) — another hormone the body secretes for a feeling of satiety.

In March 2024, Wegovy was approved by the FDA for reducing the risks of cardiac attacks and strokes in overweight patients. In a clinical trial, Wegovy reduced cardiovascular risk by 20 per cent. Another clinical trial proved semiglutide reduces chronic kidney disease (CKD) risk by 24 per cent. Almost 40 per cent diabetics also suffer from CKD. The European Medicine Agency (EMA) has allowed a lab update to include this benefit, and the FDA is likely to make a decision soon.

Novo Nordisk is conducting trial for CariSema, a combination of cagrilintide and semaglutide (Redefine 1) in December 2024. In a 68-week trial, the drug showed superior weight loss as compared to a placebo. The full results are expected in 2025.

Zepbound of Eli Lilly is the first prescription drug approved for adults with obesity and moderate-to-severe sleep apnea, (OSA). OSA was removed in 50 per cent participants.

Lilly conducted a trial in August 2024 of tirzepatide in heart failure patients with preserved ejection factor and obesity. Lilly expects a regulatory approval soon.

Lilly’s another trial (in June 2024) of tirzepatide was conducted for adults with MASH, a fatty lever disease that leads to inflammation. Half the patients have shown improvements.

In July 2024, an expert committee of India’s drug regulator has approved tirzepatide for chronic weight management with a kidney condition.

Lilly is expected to launch Mounjaro in India in 2025. Novo Nordisk is also expected to launch its weight-loss drugs in new year.

GLP-1 drugs will be introduced by Indian manufacturers, as these go off-the-patent. Some companies have started bioequivalence trials to prove the efficacy of these generic versions.

Immunotherapy is less invasive and more effective treatment for cancer. It trains the body’s immune system to recognize and attack cancer cells.

CAR-T cell therapy has been launched in India in 2024. India has developed NexCAR-19 in collaboration of industry and academy.

A patient’s T-cells are taken. T-cells are a type of white blood cells that fight cancer. T-cells are modified to recognize and destroy cancer cells. NexCAR19 treats B-cell lymphomas and B-acute lymphoblastic leukemia (in patients aged 15 and older).

NexCAR19 is the most affordable CAR-T therapy. Immuneel, a startup in Bangalore supported by Biocon and academia of Columbia University are conducting clinical trials for CAR-T therapy called IMAGINE.

Dr. Reddy’s Lab also published CAR-T therapy results of the clinical trial. It shows positive results in multiple myeloma, a type of blood cancer.

Another approach to cancer treatment is vaccination. Here personalized cancer vaccines are developed. They use mRNA technology. NHS in England is conducting trials for bowel cancer. A patient’s tumor is analyzed to detect unique mutations. It helps the immune system target the cancer more effectively. The trial will cover other cancers in future e.g. pancreatic and lung cancer in 2026.

HPV vaccination campaign has been launched in India in 2025. It protects the patients from cervical, anal and vaginal cancers.

It is still a moot point whether GLP-1s such as Novo Nordisk’s Wegovy and Eli Lily’s Zepbound are medical breakthroughs or a short-cut method to reduce weight. There is sufficient data on the health benefits of these drugs beyond reducing weight — they mitigate diabetes, kidney failure, sleep apnea and heart disease. Most users appreciate their immense social value.

GLP-1s have changed the treatment protocol of obesity but there are issues of ethics for the medical professionals — who should take them?

There is a huge market for both Wegovy and Zepbound. The FDA has approved these drugs on the basis of body mass — a body mass index of 30 or more are eligible. Those with a body mass index of 27 or higher with a weight-related condition such as high BP or sleep apnea can take them. Some 57 million Americans satisfy these criteria. They are of the working age group. Some 14 million retirement age Americans satisfy these criteria. These drugs, once commenced, are to be taken for life.

The issue is whether anyone satisfying these criteria should be prescribed these drugs. Medical professionals differ. Some are perfectly healthy at the BMI specified. And some have weight-related complications. There should be a distinction between these two.

Another issue is the shortages of these drugs and their high prices. Medical professionals will have to prioritize patients who should be prescribed these anti-obesity drugs.

The approach to GLP-1s should be evidence-based. Obesity experts will release a report in 2025 to this end. The criteria to define clinical obesity should be decided. At times, excess weight raises the risk of other health issues. Doctors should be able to identify who have the true disease of obesity (they must be treated early), and those who have excess 10-20 pounds, crossed BMI threshold and yet are healthy.

Users of GLP-1s gain weight if the ongoing treatment is discontinued. This is also true for those who have weight-related issues. The issue therefore, is when is someone’s weight loss enough. The doctors advise them to continue and to stick to lifestyle changes. The risk is that such long term continuation makes you lose muscle along with fat. This influences nutritional requirement.

There is no consensus among the medical professionals to manage obesity using drugs. Is BMI the right metric? Or should it be the weight-to-height ratio? Should the health markers be considered? The picture becomes more complicated because of the health benefits of GLP-1s that accrue, regardless of the quantum of weight loss

Doctors have standardized treatment goals for cholesterol, sugar and BP since the criteria are now well-settled. In case of obesity drugs, patients have strong opinions on these criteria despite evidence of what is good and bad.

Novo Nordisk, Denmark’s pharma company, producing anti-diabetic insulin, Ozempic and Wegovy entered the anti-obesity market worth billions of dollars and witnessed a spectacular rise. It is equivalent to the rise of Nokia of Finland in the cell phone market, till it was displaced by other smart phones including Apple’s iPhone. Can Novo go the Nokia way?

Novo’s share declined by 21 per cent. There was a trial of next generation anti-obesity drug CagriSema. The drug fell short of 25 per cent weight loss that was predicted. The recorded weight loss hovered between 20.4 per cent to 22.7 per cent over a period of 68 weeks. It is close to rival Eli Lily’s Zepbound shot. However, it dilutes the European firm’s market leadership.

Eli Lily’s Zepbound has a 40 per cent market share in the USA. It acquired this share in less than a year. Lily is trying an experimental compound retarturide that showed a 24 per cent weight loss in a similar trial. There are other competitors such as Amgen and Pfizer who are eyeing this market. There is patent expiry of Wegovy in the early 2030s. Some of these developments may affect Novo sales by 2030.

The developments are in the interest of the consumer and yet these are quite humbling for Europe. Goldman Sachs group has devised an acronym for mega-cap — GRANOLAS capturing top companies with a competitive advantage — GSK, Roche, Astra Zeneca, Novo Nordisk, Nestle, Novartis, L’Oreal, LVMH, ASML and SAP.

There are spillover effects of Novo’s boom — it employs 32000 people in Demark, there are new manufacturing sites in France, and substantial exports. These could become vulnerable and could also affect Denmark’s currency. Still Novo is a well-diversified company, and we cannot imagine a Nokia-like effect. There could be a different ending this time.

Noam Shazeer, an AI/ML scientist and research worker, was with Google for 20 years, and then walked away to set up Character.AI. He has returned to Google at an astronomical price of $ 2.7 billion or Rs 23000 crore in Indian currency.

Noam Shazeer is 48 now. He is extraordinarily brilliant and a trailblazer. He lives in Silicon Valley, California, which is the beating heart of the international tech revolution. Noam’s coding skills are dazzling. At Duke, he learnt computer science and mathematics. The subject gave him a deep understanding of algorithms, data structures and mathematical modeling. He joins Google in early 2000s, after his academics are over.

He designed Google’s spelling correction system. He elevated it to an unprecedented accuracy. Spelling snafus could be solved by mathematics, owing to geniuses such as Noam.

In 2016, he developed sparsely gated mixture of experts. He designed personally the multi-head attention of Transformer architecture. He facilitated its implementation by coding. He was a major contributor to Google’s LaMDA dialogue system in 2019. This project was led by Daniel De Freitas (who later cofounded Character.AI with Noam).

Noam Shazeer had a conducive home environment. His father was a polyglot engineer. His mother valued education and curiosity. She was a homemaker.

De Fritas, along with his mother, lived in various countries such as Venezuela and the UK. Ultimately the family settled in Florida along with his adoptive father. His upbringing was multi-cultural. It paved the way for strong foundation in natural language processing (NLP).

While at Google, Noam and Freitas developed Meena — an advanced chatbot with conversational skills. Google did not push this project forward. Disillusioned, Noam and Frietas left Google. They co-founded Character.AI, a platform for generative AI. The conversations could be carried out with fictional characters and historical figures. They could be conducted with customized personas. Users could do role playing or could conduct fantasy-based conversations. It led to complications, some of which were legal.

There is a twist in the story. Noam Shazeer and Freitas — both returned to Google in August 2024. The return of the prodigal sons. Shazeer reportedly received a staggering $ 2.7 billion to come into Google’s fold.

Google welcomed back Noam Shazeer, a preeminent researcher in ML who is joined DeepMind. Freitas too returns to the search titan. It is monumental homecoming.

In this high-stake world of AI, there are pushes and pulls. They returned on their own terms. The future is wide open.

o3 from OpenAI has achieved human-level results on a test designed to measure general intelligence. The evaluation test is called ARC-AGI on which the previous AI systems could not cross 55 per cent, though the threshold set for human-level results was set at 85 per cent. The test is a tough mathematics test. Apparently, OpenAI has achieved a significant step towards the goal.

Let us understand ARC-AGI test. It is a test of AI system’s sample efficiency in adapting to something new. An AI system is tested against new examples of a novel situation which the system has to figure out.

ChatGPT is not very sample efficient, as it is pretrained on millions of examples of human text and constructs probabilities rules about combinations of words most likely to follow. It is good at common tasks but not so good or bad at uncommon tasks — it is exposed to less data or fewer such samples of these tasks.

The ability to generalize in essence means to solve previously unknown or novel problems from limited samples of data. This is an important component of general intelligence.

The ARC-AGI benchmark tests for sample efficient adaptation. It uses little grid-square problems. The AI needs to figure out the pattern that turns grid on the left into the grid on the right. Each question poses three illustrations to learn from. The AI system needs to figure out the rules that generalize from these three examples to the fourth.

It is akin to IQ tests in some school tests and competitive examinations.

OpenAI’s o3 model is highly adaptable and it has done it. In detecting the pattern, there should not be unnecessary assumptions. At the same time, we should not be more specific than it is needed. The weakest rules, in theory, are the ones that do what you want to do. If you can identify these, it maximizes your ability to adapt to new situations.

Weaker rules are the ones phrased in simpler statements.

OpenAI may not have optimized o3 model to find weak rules. However, to succeed at ARC-AGI, it is necessary to find them.

OpenAI began with a general-purpose version of the o3 model, and then trained it specifically for the ARC-AGI test.

Francois Chollet who designed the benchmark believes o3 searches through different ‘chains of thought’ to solve the task. It then chooses the ‘best’ according to some loosely defined rule, or heuristic’.

The chains of thought could be considered as programme that aligns with the examples.

Several different programmes are generated. The heuristic could be ‘chose the weakest’ or ‘choose the simplest’.

Is this what is closer to AGI? If the modus operandi of the model is what has been described above, it would not be better than the previous models. Its learning from language generalization is not suitable. A more generalizable ‘chain of thought’ could be seen through the extra steps of training a heuristic specialized to this test.

Much of what has gone into the making of o3 remains unknown. Its exposure is still limited to a select audience. Its understanding will require extensive work. It should be seen how often it succeeds and how often it fails. Let it hit the market. We will then know whether it is as adaptable as an average human being. If it is, it can be further improved to achieve accelerated intelligence. It is to be governed too by certain framework. If it is not adaptable, we should call the test result impressive, but life goes on as usual.

Though cinema is an art form, it is also a major investment and business. To make cinema successful, the makers use a several different elements together to make a coherent story. It is peppered by music and choreographed dance sequences contributing a major chunk of run time.

There are emotional scenes between parents and children, raunchy dance numbers called item numbers, romance sub-plots and action sequences. At the end, all these elements fall together to form a medley of flavours. It is akin to how different spices are combined together in Indian cooking. Such films are thus called masala films.

Masala films were inspired by historical epics such as Mahabharata, Ramayana. There are side characters, flashbacks and sub-plots. Ancient dramas too inspire masala films — Abhigyanshakuntalam of Kalidasa where song and dance create stylized spectacles. There is emotional response from the audience.

Indian folk theatre also influences masala films, e.g. Ramleelas. The Parsi theatre also influences masala films — crude humour, melodious songs and dazzling statecraft.

Yaadon Ki Barat (1973) was the first pioneering masala Hindi film. It was directed by Nasir Husain. Salim Javed then created so many masala scripts — Sholay (1975), Deewar (1975), Don (1978). These were remade in other languages.

Masala movies cater to a wider section of population. As there are different elements, there is going to be some element that appeals to you. These are the movies for family audiences. Even the social message delivered by such movies reach a wider audience.

Many masala movies are being made in the south. The recent film of Allu Arjun Pushpa 2 is a big hit; though originally made for Telugu audience, it attracts a wider Hindi audience.

After the economic liberalization of 1991, there was a thematic change in Hindi movies. They started making movies for a large Indian diaspora living abroad, along with the local audience. The NRIs affected the storyline. The first such change was seen in 1995 movie Dilwale Dulhania Le Jayenge (DDLJ). There was a change in sensitivities due to western exposure, and a special term multi-plex movies described such niche movies. The urban audience was the primary attraction.

This two does not fade away the trend of masala movies completely. We now see a mix of multi-plex films and masala movies.