Man-machine merge when machines imitate human being and better them in some respects. It is called a state of singularity — a point when machines become smarter than human beings. It results into Artificial General Intelligence or another name for human-level intelligence. It is expected that machines may match human intelligence by 2029.

Even before the advent of computers, a possibility of thinking and decision making by machines was considered in the 17th century. The maxim was, ‘I think and therefore I am.’

Turing is said to be the pioneer of AI who floats the idea of thinking machines. He developed the idea of the Turing Test. It decides whether a computer is capable to think or not. McCarthy coined the term AI to define this ability in 1956.

The first natural language processing computer Eliza was developed in 1966 at MIT’s AI Lab. It was a forerunner of the present day digital assistants such as Siri and Alexa.

In the 1990’s, Lee published how HTTP worked which led to world wide web. We assigned the name Internet to this. Internet ultimately led us to big data, which is fuel for AI.

Deep Blue developed by IBM (1999) defeated Garry Kasparov in chess. Watson a computer engine from IBM overshadowed humans in TV game show Jeopardy! In 2012, computers could recognise cats when trained to do so.

In 2018, we have developed autonomous cars. In 2016, AlphaGo from Deep Mind defeated world Go champion Lee Sodol.

AI is likely to penetrate every commercial sector. Customised products could be developed by using AI in banking and insurance. Crypto tokens can be a medium of exchange. Self aware and self replicating software systems could emerge.

The Google play store has a ‘Spam and Minimum Functionality Policy’. It can take down apps for repetitive content, among other things. There could be violations of copying content or value or creation of multiple apps with highly similar functionality content and user experience.

There are organisations who sell the source code of the apps. There is a market place called CodeCanyon for the buying and selling of scripts, entire apps and components for a variety of languages and frameworks. Buying source codes of popular apps is quite a common practice everywhere in the world. There are at least 500 developers selling source code for popular apps. But just source code is not enough. There are clones of popular apps and there are clones of clones. These clones have no well-defined data usage and privacy policies. Source code apps are bought cheap. They could be slightly modified. They are then re-published on app stores. People marketing such apps through ads and/or in-app purchases. Apparently, these apps are similar to those developed by larger companies. These fast apps could hurt the consumers if sensitive data is compromised. Many are counterfeit or fake apps.

Google did deep learning research using a large data set of retinal images to diagnose diabetic retinopathy which leads to blindness. AI could prove transformative in healthcare. Google has developed an algorithm which is trained by a data set of 1,28,000 images. Each image has been reviewed by a 3-7 expert ophthalmologists. These are fundus images. The symptoms could be micro-aneurysms, haemorrhages, hard exudates etc. AI detects referable diabetic retinopathy (DR). It is possible to do this fast and in greater numbers. Early diagnosis is preventive. The software was tested using 12000 images. The performance rating was 0.95, the highest being 1.The ophthalmologists rating was 0.91. Google now is focusing on 3D retinal images. It may point out cases which doctors would have missed. Instead of a doctor alone or algo-alone, it is better to have an algo complementing a doctor.

Such algos can be used to detect cardio-vascular disease.

The govt has set up four centres for promoting Industry 4.0. Many more such centres are required. Manufacturing should become smart and intelligent. The process industries have already done a lot of automation and are among the first adopters of AI, e.g. chemicals, pharma and auto. The rest of the industry is adopting digitisation. They would like to unlock the data present in their systems to facilitate intelligence.

Digitisation

It is the process of capturing all the content and data of a company in digital form in a computer. It is stored to optimise for AI. This is followed by automation. Then comes AI.

AI involves the stages of visualisation, data analysis, and prediction of outcomes.

ML algorithms should work on the data for historical analysis, running through it as many times as possible or iterating. The algorithm learns, and optimises itself to take decisions in the future.

Central Data Repository

The govt proposes to set up central data repository to connect all the elements of India’s fiduciary structure — centre, states, cities, ministries. It will be a planned network. It will make data available for AI. This will be useful for urban planning and making cities smarter.

Healthcare, education sector, logistics and transportation plan to do this.

Post-covid film making has been allowed after 10 weeks of shut-down. There is going to be lot of adaptation in film, TV and OTT shooting. Casting could be done remotely. Scenes would not have crowds and many artists in one frame. Junior artists would have to wait till they again become part of the project. The idea is to observe social or physical distancing. Screenplays would adapt to the requirement of post-Covid shooting. Crews will be cut to a third of usual size. Make up of artists other than the lead artists would be managed by the artists themselves. Everything from transport to budgeting has to be reworked from scratch. Every shoot could be considered like an outdoor shoot where people do not return home at the end of the day. Locales which are not so busy could be chosen. First TV episodes would be resumed to test the waters. Film shooting can resume later. Sets may not have ACs. Some studios would commence shoots of ads and music videos. Later incomplete films would be completed. All departments would be affected by downsizing, say art, setting lights and camera. Stunt artists would not be employed immediately. Film units would have to become more efficient to work with lesser manpower. Directors would have to choose new themes to cater to the public mood, rather than escapist cinema. There would not be songs with 100 dancers. Campus stories would be shelved. OTT would be used to release many movies. Film budgets would shrink. There could be cost correction in all departments. Many films would be made on shoe-string budgets. Multiplexes would have to be careful about sanitisation and house keeping. Audience could not be jam packed. There could be vacant seats to maintain social distancing. Occupancy would be less initially. Though streaming movies is an option, collective viewing experience of theatres would survive.

In May 2020, the value of bitcoin mining halved to 6.25. There are 21 million bitcoins in total, and 87.5% of them have been mined. Halving means reduction in the cost of mining every time miners hit a specific number of blocks. The finding more blocks and clearing them is mining. The value of bitcoins shoot up. It was considered a digital gold rush. More people involve in mining. Every time 2.10 lac blocks get cleared, the value halves. At first 2.10 lac blocks, the value is 25 bitcoins. At 4.2 lac blocks, it is 12.5 bitcoins. At 6.3 lac blocks, it is 6.25 bitcoins. The value reduced till 2140, when it becomes zero. That is the point when the transaction fees would increase.

As there is block difficulty, progressively it is difficult to mine a block. This is the built-in safety value. It protects it from too much mining.

It is challenging to mine bitcoins over a period of time — more expensive equipment and servers to do so. The cost rise. Amateurs quit the game. At the same time, the price of bitcoin rises. Bitcoin has a limitation — it can store only 4MB data. Thus unless the blockchains of other countries interact, its utility will not be complete.

IBM has exited facial recognition technology business. Microsoft is reluctant to sell facial recognition technology to police. Amazon put a moratorium of one year on selling ‘Recoknition’ to law enforcement agencies.

The technology is likely to be measured for mass surveillance and racial profiling. It could violate basic human rights and freedoms. It is too dangerous to be used for law enforcement purposes right now.

The governments are slowing down or stopping altogether uses of facial recognition in public sphere.

Any government which could identify every face in public everywhere all the time deals a death blow to freedom.

Government must protect public order but must subject the privacy invading technologies to judicial review. It must introduce legislation to formalise what transpires in judicial review.

If every face is recognised, it will be difficult to have political dissent.

Blockchain is a database. It has advantages such as decentralisation and immutability. It is highly secure. It is a list of records or transactions similar to a ledger, that keeps growing as more entries are added. Records stored in the database may be made visible to the stakeholders without the risk of alteration. It is said to be highly secure as hackers can no longer attack just one computer to change any records.

Blockchain has found application in delivering vaccines on time. Every touch point of the vaccine’s journey, starting from manufacturing to the child to whom it is to be administered, critical information about it such as temperature, humidity, chain of custody and location are recorded on a blockchain ledger. The temper-proof data is accessible to all the stakeholders to ensure transparency.

In chit funds, money from subscribers is pooled at definite time intervals. There is an auction to borrow pooled amount. The money is given to the lowest bidder or winner. The chit fund handler also gets a collateral from the winner. By blockchain, each record,is digitally maintained and can be traced by the regulator. It brings sanctity to the whole process. Multiple digital ledgers can be created.

As you are aware, social media such as Google and Facebook carry the news articles and high quality stories. As the audience gets free access, they avoid the media from which the stories are sourced.

Gathering news, editing it and validating it, and putting it in perspective and context is a costly business. Who pays the costs? Obviously, the advertisers. In return, they reach wider audiences. This is the model of the traditional media. Had there been no advertising, consumers would have to pay far more for their daily dose of news and its analysis.

In today’s world, Google and Facebook duopoly get most of the advertising, leaving meagre revenue for traditional media companies. It is difficult for them to invest in newsrooms and generate the content ultimately to benefit the aggregators such as Google and Facebook.

Ideally, the tech companies must share the advertising revenuue with those who generate content. Though aggregators do perform a service, that by itself cannot entitle them to the total advertising revenue generated by the news content.

Increasing number of people consume news online. It may be a demographic change or habit change. Such a change starves the traditional media of advertising or subscription revenue. This could kill the news business itself.

Google has started thinking about this and is considering to part with some of its earnings with the media companies.