The images are wonderful. The majority of the ideas have come true, but even for the ones that have come true, their achievement was not at all in the way that they were visualized in these images.
Image Credit - Morlet's Blog - Police X-Ray Surveillance Machine
I am going by memory here, but I believe that Robert A. Heinlein, in an essay in the Expanded Universe collection, talks about how science fiction authors tend to miss the mark because they are insufficiently imaginative. Add futurists to that list - there was no such profession when Heinlein wrote that essay originally in 1950, and I don't remember whether he mentions them in his 1980 update.
Technological progress must resemble what I imagine biological evolution to be like -- a long period of relatively small adaptations for existing conditions which don't change, but then a sudden mad scramble for adaptation to a radical change in the environment. Think more or less stable conditions for a hundred million years to allow dinosaurs to evolve and adapt to almost all ecological niches, and then being blotted out by radical environmental change after the Chicxulub asteroid strike. Having seen both Tyrannosaurus Rex and Triceratops, it might have been difficult to predict polar bears and bison.
In just the same way, in 1900, it wasn't obvious about the various revolutions caused by cheap internal combustion engines, air travel, rocket flight, space travel, computers, genetics -- the list goes on.
Here in no particular order is my short, incomplete list of problems we currently face, and possible solutions. It is no great prediction that from the year 2100, readers can look back on this list with the same sort of amusement that we in the post-2000 era look back on the predictions of 1900. As with the 1900 predictions, the flaw in these will be the unknown and unanticipated discoveries (or problems) that derail these and cause the outcome to be different.
GLOBAL WARMING - at the end of the 1800s, there was a hard limit to how big a city could get, because they were already running into the problem of how to get rid of all the horse manure that a large city would create. The analogy to our present day problem with rising carbon dioxide levels due to industrialization and transportation are obvious. How does this get fixed? Prediction - a range of technologies, including bioengineered forests (think kudzu genes implanted in sequoias) to absorb atmospheric carbon dioxide. In addition, there will be technology applied to remove carbon from the atmosphere and store it.
ENERGY - The industrialized world demands an enormous amount of energy. Energy needs will increase as Asia and Africa expand their consumption to levels that approach North America or Europe. Energy sources such as coal and oil release too much carbon. Nuclear reactors create dangerous waste. Space power satellites are expensive. Prediction - fusion reactors based on compression or pinch effects -- for example, the Bussard concept.
CONFLICT AND GOVERNANCE - In the shorter to medium term, this is also about creating a way to equitably share resources - something that has more often than not resulted in conflicts. Thomas Barnett's TED talk examines the way that conflicts (for example over energy and resources) could evolve, and the possible international solutions to these problems. Prediction - armed conflict won't go away, but the scope and intensity will drop as the main international players understand that war is bad for business.
SPACE - GROUND TO ORBIT - Rockets were all right as a first step for getting things into space, but over the long term, they are dangerous, polluting monsters. Prediction - Multistage scramject launchers will be developed, but will give way to an orbital elevator. If the elevator is based on some variation on carbon nanotube technology, several of these might be a good place to use up the carbon that needs to be removed from the atmosphere.
SPACE - SOLAR SYSTEM - The problem with the rockets that are being built is that rockets fundamentally can't lift very much into space -- most of their energy output is devoted to moving their own fuel. In order to get anywhere interesting with anything of a decent size, this constraint has to be removed. Scramjet launchers will help, so will an orbital elevator. Prediction (see also ENERGY above): A Bussard-type fusion reactor (or something similar) will turn transit times of years to weeks or months, putting the solar system out to the Kuiper Belt in range of manned exploration.
SPACE - BEYOND - There are mathematical models for how to get something to move faster than light, but no one has any idea on how to implement these concepts as buildable engineering. Prediction - by 2100, we will have a better understanding of spacetime, and will be able to create devices to allow local manipulation of gravity. Faster than light travel will remain elusive.
GENETICS - The molecular basis for life was not understood in 1900, but by 2000, the genome of human beings has been determined, as well as that of some other species. Prediction - by 2100, genetic data will be so readily available that researchers will have a much better understanding of the evolutionary heritage of the majority of creatures on this planet. Some extinct species will be recreated, although a "Jurassic Park" scenario will not be achievable. However, greater understanding of cloning technologies will enable organs and other spare parts to be grown as needed. Meat will be grown in vats rather than harvested from food animals.
NANOTECH - Very primitive nanotech devices have been created, but no one really knows yet how to do really sophisticated, sexy stuff. Prediction: the nanotech revolution will go hand in hand with developments in genetics and molecular biology. There will be something like a genetic code for self-assembling machines. These machines will become ubiquitous in the environment.
ARTIFICIAL INTELLIGENCE - Even as late as the 1940s, there was a belief that computing devices would be mechanical. The computer revolution first of all needed to go electronic, and after that, took 40 years to get from the first mainframes to the internet. Having gotten to this point, it shows no sign of slowing down. If anything, the pace of development is increasing. Prediction - this one is the joker in the deck, because it affects everything else. As computers have gotten smarter over the years, we have used them as tools to allow us as individuals or groups to do more. Computer systems will continue to increase in power, and well before 2100, a computer (or computer network) will be able to match the power of a human brain. But it won't stop there - computer capabilities will continue to expand. We will share our world with AI entities who will help us to organize our data, our work, and our lives. This will affect all other human endeavours.
NEW PROBLEMS - Sure as anything, there will be unintended consequences -- stuff we didn't think of. Hopefully not fatal.