This week, Google's DeepMind AlphaGo AI beat the world's number one Go player Ke Jie - but what is Artificial Intelligence?

Artificial Intelligence (or AI) is the simulation of human intelligence by machines. It's a buzzword that's really entered the mainstream in recent times - but its origins began in World War 2.

British mathematician Alan Turing and neurologist Grey Walter first tackled the idea of intelligent machines some 70 years ago at the Ratio Club, an influential dining society for biologists and engineers.

While Grey built some of the first ever robots, Alan invented the famous "Turing Test". Basically, if a machine could fool someone into thinking they were talking to another person, then it would pass the test.

But it wasn't until 1956 when the term "Artificial Intelligence" was coined by computer scientist John McCarthy.

There are a lot of different benchmarks with regards to building AI systems. They fall into three broad categories: Strong AI, Weak AI and something in-between the two.

Strong AI aims to genuinely simulate human reasoning. A strong AI system would beat the Turing Test in a mechanical heartbeat.

A strong AI system can think like a human, and perform tasks on it own. There's no need for a human to manually enter a task that it wants the strong AI system to do. They can make decisions on the spot.

No strong AI machines exist in the real world, but Kryten from Red Dwarf is a sci-fi example of such a system.

He's self-aware, can learn new skills, perform tasks independently and shows emotions, as the below video shows. Yes, you would be able to distinguish Kryten from a real human by looking him, but not if you were chatting to him without seeing him in a computer chatroom, for example.

Weak AI is a computer system that acts like a human but does not understand how humans think. ​It is non-sentient AI, which is focused on one narrow task.

Google's aforementioned DeepMind AlphaGo AI is a good example. ​Although it beat the world's number one human Go player, it did not play in the same way that humans do.* It's the same for weak AI systems like Siri, or your car's sat nav. 

Then, we have something in between the two. Or In-between AI, as I'll call it now. These systems may not perfectly model the human mind - but they use human reasoning as a guide.

The IBM Watson AI machine is in the in-between category. It's, essentially, a chatbot that can answer your questions using natural language.

But I've had complaints about the lack of Lego in the last couple of posts, so I've got to use Ironman's cyber-butler J.A.R.V.I.S. as an example of in-between AI.

J.A.R.V.I.S. was originally a bit like Siri - a highly advanced language-based interface. Over the years, Tony Stark updated J.A.R.V.I.S. so it appears that the system has a mind of his own.

We don't know the exact details as to how J.A.R.V.I.S. works - but he seems to build up evidence by looking at thousands of pieces of information to give Tony a valid conclusion when asked a question. And he talks in a similar manner to a human - which is a true example of inbetween-AI.

In other words, J.A.R.V.I.S. (Just A Rather Very Intelligent System) makes decisions as a human would. He looks for patterns in the evidence to reach a conclusion by weighing up different elements.

Things aren't black or white to J.A.R.V.I.S. - just are they aren't for human intelligence. But J.A.R.V.I.S. still needs a human to function, so it's not yet** a strong AI system.

Some researchers have argued that the Turing test was passed by a chatbot in June 2014 called Eugene Goostman that fooled people into believing it was a 13-year-old Ukrainian boy.

But some critics claim the test was not long enough and the fact that it was talking in its non-native English language meant the machine had an unfair advantage.

Either way, we're a long way from building a self-aware machine that can call us a bunch of smegheads.

Here's a more in-depth discussion on the strong and weak types of AI (also known as general and narrow). Here's a great video detailing the differences between strong and weak AI too:

Here's a longer video from John Searle, where spoke about the philosophy of mind and the potential for consciousness in artificial intelligence at Google's Singularity Network.

The future of AI is a fascinating topic, with many tech experts disagreeing on its future directions. It's also interesting to consider how a computer system could evolve into a strong AI system.

To quote my favourite series 4000 Mechanoid: "Please sir, give me some credit. I am not the one-dimensional cleaning droid I was once was; I've evolved into something far more complex and multi-layered, and if I may so say so, superior."

​And here's a final, fun example of the difference between a Strong AI system (Kryten) and a Weak AI system (the talking toaster):

* AlphaGo is a step up compared to other weak AI systems like IBM’s master chess beater Deep Blue. While other weak AI systems rely solely on constructing a search tree over all possible positions, this wouldn't work for a game of Go where players take turns to place black or white stones on a board, capture each other's stones and try to get the most amount of space on the board.

Go is a simple game to describe - but very much more complex than a game like chess. So, AlphaGo used an advanced tree search combined with functionality informed by the behaviour of the brain - called deep neural networks. ​I'd be tempted to say AlphaGo could be classified as in-between AI - but the experts disagree.

​​** I'm describing an early version of J.A.R.V.I.S here - not the subsequent versions seen in the Marvel world, which are closer to Strong AI, particularly F.R.I.D.A.Y. - the AI system seen in the Age of Ultron film. 

The Enigma machine is instantly synonymous with the codebreaking work done around the Second World War. But you may not have heard of the Lorenz machine - which was a much tougher encryption nut to crack and was responsible for passing the most important messages between German high command.

Enigma only had three to four wheels, which were used to scramble messages being passed between the German forces. Lorenz had 12 wheels. 

To put this into context, there were 159 million million million possible settings to choose from when you're setting up an Enigma machine.

​For Lorenz, there were more than 1,000 million million million million million million million million million million million million million million million million million million million million million million million million million million million million total number of combinations.

Or 10^170.

​For an encrypted message to be successfully passed between two Lorenz machines, they both had to have the same wheel settings. 

Lorenz was attached to a teleprinter and the message  to be encrypted or decrypted came in ​as a long stream of paper imprinted with Baudot teleprinter code.

The Baudot code is a character code where each letter is represented by five bits: either a + or a -.

So, the letter A could be "+----" and the letter Z "+---+", for example.

What the Lorenz machine would do is add on a random letter to the one being passed in the message.

For example, let's say we want to pass the letter A as the message between two Lorenz machines. The Lorenz may generate the letter Z as the key.
Lorenz code = message + key.
So, to encrypt the message, you just have to add each bit (the + or -) of the letter 'A' with each bit of its key 'Z'. The resulting letter will be encrypted.
If two symbols are the same, then we generate a "+" bit in the encrypted code. If they are different then we get a "-".
In other words: + + + = + or - + - = +
And: + + - = - or - + + = -.
So, if we go back to our example where we are encrypting a letter A with a key of Z:
Message A: + - - - -
Key           Z: + - - - +
Code        D: + + + + -
So, using these rules the letter 'A' is encrypted as a letter 'D' using the key 'Z'.
Here's the clever part. If you receive the letter 'D' and know the key is the letter 'Z' - you can add the bits together again to get the letter A out. In other words, the system works both ways so you can encrypt and decrypt your message - if you know the key.
The key was produced using the combination of the two sets of five wheels on the Lorenz machine, whose movement were controlled using two motor wheels.
So, using these rules the letter 'A' is encrypted as a letter 'D' using the key "Z'.

Here's the clever part. If you receive the letter 'D' and know the key is the letter 'Z' - you can add the bits together again to get the letter A out. In other words, the system works both ways so you can encrypt and decrypt your message - if you know the key.

The key was produced using the combination of the two sets of five wheels on the Lorenz machine, whose movement were controlled using two motor wheels.

The real breakthrough for Lorenz came due to operator error. The same Lorenz-encoded message was sent twice - but the operator was so annoyed that they had to retype the 4,000 character message, they put in some abbreviations.

This meant that the codebreakers at Bletchley had two copies of the same message, with a few alterations and using the same key.

So, if you add the code together, then two messages can be inferred and the code can be broken.

You can get the all-important key.

Mathematician Bill Tutte was given the key to see if he could find any patterns to identify the length of the key. This involved him writing out the key in long rows.

When we wrote the key out in rows of 41, a pattern started to appear. This meant the first wheel in the Lorenz machine had a period of 41.

The pattern wasn't perfect and this left Bill to deduce that there was another wheel that sometimes moved. He was right - he'd worked out that Lorenz relied on two sets of wheels.

Using this approach, Bill and other codebreakers could deduce the layout of the Lorenz machine - without ever seeing the machine itself.

Bill Tutte also came up with a procedure to work out the initial settings on the Lorenz machine. Although this procedure worked in principle, it would have taken too long to carry out and decipher messages by hand.

So, the process was automated and the world's first computer Colossus came into being. 

More on Colossus next week.

For a more in-depth explanation of the Lorenz Cipher and how it was broken, click here. If you want to play with a replica of the Lorenz cipher system - check out this amazing Virtual Lorenz machine from the UK's National Museum of Computing (TNMC).​

And here's a quick video covering the "unbreakable" Lorenz code:

Last Sunday, I visited Bletchley Park for a symposium and the unveiling of an exhibition on mathematician Bill Tutte on what would have been his 100th birthday.

You may not have heard of Bill Tutte but (in what was later called “the greatest intellectual achievement of the war") Tutte helped to crack the Lorenz code that the German forces used to encrypt and communicate messages during World War II - without ever actually seeing a working model.

His work extends far beyond his days at Bletchley. Bill was educated at Cheveley Village School in Cambridgeshire, before going to Trinity College, Cambridge, in 1935. He studied Natural Science and specialised in Chemistry - but his childhood love of mathematics grew during his time at Cambridge.

Bill became close friends with three mathematics students: Leonard Brooks, Cedric Smith and Arthur Stone, and the foursome spent their time solving mathematical problems. They were particularly attracted to a problem known as "Squaring the Square" - a simple puzzle where you are tasked with dividing a square into smaller squares of different sizes.

Simple as it may sound, it was assumed at the time that this puzzle could not be solved. But the foursome cracked it by discovering an unexpected link between electrical circuits and mathematics. They were pipped to the post by Roland Sprague, a German mathematician, who published the solution - but not the theory behind it.

Speaking at the symposium, Claire Butterfield from the Bill Tutte Memorial Fund, said: "Mathematics was where his true passion lay and his reputation as a problem solver meant he was interviewed at Bletchley."

In May 1941, Bill arrived at Bletchley Park and was put to work in the research section. He initially worked on the Italian Naval Cipher, before he was introduced to the Lorenz Cipher.

The Lorenz system was more advanced than the Enigma machine - and the British knew very little about how it worked. So, Bill and the team were faced with a harder problem with less information to solve it.

But he did solve it - in about six months. Bill's approach was to try to work out how the cipher wheels on the Lorenz machine (known to the Allies as Tunny) worked by writing down on squared paper the first dot or cross of every character of the cipher text. 

Just like the countless mathematical problems he'd worked on during his life, he was searching for a pattern. Bill once said: "I do not think I had much faith in this procedure, but I thought it best to seem busy."

Using this approach (and a security blunder by a German operator where two versions of the same message were sent with sloppy changes), Bill and the other members of the Research Station worked out the structure and movement of the wheels on Tunny. 

Dr David Kenyon, a research historian at Bletchley Park, said: "Tutte's contribution was a turning point in what Bletchley Park was able to achieve during World War Two. The Allies' understanding of the German plans in France prior to D-Day is very significantly based on Fish [Lorenz] intercepts rather than Enigma. Had they not had this intelligence, their understanding would have been much weaker."

To find out more about Bill's life in codebreaking, click here.

After the war, Bill moved back to Cambridge to finish his PhD. His love of mathematics continued and he had a real interest in a phenomenon known as "Graph Theory".

His PhD advisor (and a fellow codebreaker) Shaun Wylie wasn't keen for Bill to conduct research in this area. Claire said: "He was recommended not to do Graph Theory as it was seen as trivial in the field of mathematics - but Bill just enjoyed solving problems."

There was no possibility for Bill to continue this work in Graph Theory at Cambridge after his PhD - so he moved to Canada where he first taught at the University of Toronto, before moving to the Univerity of Waterloo, where he held the position of Professor of Combinatorics and Optimisation.

"He had the luxury of teaching subjects that were close to his heart," Claire added. "His lectures had an element of storytelling about them and he would also teach local children to play chess over milk and cookies."

As idyllic as Bill's life in Canada sounds, his professional career had just as many highs. He received the Order of Canada and was elected a Fellow of the Royal Society of Canada, and a Fellow of the Royal Society of London.

His work on Graph Theory transformed this field into one of the most important areas of mathematics. Bill's work in Graph Theory led to some of the key mathematical developments that have shaped the internet today, including the science behind search engines.

​The thread that pulls Bill's extraordinary life together is that he did all of this because he followed his heart.

He had no agenda, no lofty ambitions - he had a love of mathematics and puzzles, and an aptitude to communicate these ideas effectively.

​That's what really struck a chord with me as I listened to the series of lectures on the life and work of Bill Tutte. Against all the advice from his peers, he worked in an area that was of little to no interest to the wider mathematical world. And against his own better judgement, he tinkered with a statistical method that meant Lorenz could be broken.

His work had a huge impact on the outcome of World War Two and the technology we rely on today. And that's really the beauty of Bill Tutte's mind. He had no ego - he just understood that from the simplest questions and concepts, the most complex machines like Lorenz can be cracked.

I'd like to think Bill may agree with my mantra that "Science is Simple" and that any work in science and technology is important and interlinked. Whether it's the musings of a child who questions why we don't hurtle off the Earth as it whizzes round the Sun, or a scientist sitting in CERN trying to find the most elusive matter in the universe with every bit of scientific kit as their disposal - there is no room for arrogance and assumptions in science.

Thank goodness there was room for Bill Tutte.

I'm going to end with a quote, beautifully written by the man himself as he introduces the seminal book on Graph Theory, Theory of Finite and Infinite Graphs by Denes Konig:

Today, I'll be visiting the wonderful Bletchley Park as they unveil a new exhibition for code breaker Bill Tutte (more on that in later blog posts, I promise). So, for this week's Sunday Science, I wanted to tackle one of WWII's most infamous machines - Enigma.

What is the Enigma machine?

At first glance, it looks like a fancy typewriter. But, when you press a letter on the keyboard then another letter flashes up.

Is it broken? No.

Enigma was used to encode messages. Essentially, it used a system where one letter is replaced by another.

It represented a new form of encryption using machines rather than codebooks or hand cyphers.

​If you opened up Enigma, you would see three wheels, known as rotors. Each rotor has 26 different positions representing the 26 different letters of the alphabet. These rotors can be taken out and their position swapped.

The inputted letter tapped on the keyboard passes through these three rotors and bounces off a reflector at the end, before passing the letter back through all three rotors in the opposite direction.

The encrypted letter flashes up on the lamp board above the keyboard.

Once the letter has flashed up, the first of the three rotors clicks round one position. This means that the output has now changed. 

So, even if you hit the same letter on the keyboard again, a different output letter would flash up.

When the first rotor has turned through all 26 positions, the second rotor clicks round. When that has gone through 26 clicks, the third rotor takes over.

Another component known as a "plugboard" swaps pairs of letters as they go into and out of the machine, adding another layer of complexity to the system.

The reflector means you can use Enigma to both encrypt and decrypt messages. It's the same process, just in reverse.

So, let's assume you have two Enigma machines in two different locations, set up with the same starting rotor positions and plugboard setup. Messages would be encrypted by one Enigma and passed by Morse code to the other Enigma, where the message would be deciphered.

It's a simple system to describe - but one that's incredibly complex to crack.

​To crack Enigma, you needed to know the starting position and the order of the three rotors, plus the set up of the plugboard.

There were 159 million million million possible settings to choose from when you're setting up an Enigma machine.

Oh, and the settings were changed every day during WWII so you were constantly fighting against the clock. You'd have 24 hours to try out 159 million million million possible settings.

So, the Germans believed Enigma was unbreakable. They were wrong.

There were a couple of pitfalls to Enigma's design, which meant thousands of possible rotor positions could be eliminated.

First, no letter would ever be encoded on itself. Hit "A" and you'll never get another "A" out. Second, certain phrases were very common in the messages. Operator sloppiness also helped - the rotors may not have been reset at the start of a new day, for example. 

The teams at Bletchley Park built their own machine to break the Enigma machine. Known as the Bombe, it further cut down on the number of combinations.

​This meant the teams at Bletchley could fight against the clock and beat Enigma.

It didn't end there. Enigma was continually updated to make it more difficult to crack over the years - and then the Germans introduced the more complex Lorenz cypher machine. This was cracked using another Bletchley machine - Colossus.

It's estimated that the work done at Bletchley Park shortened the Second World War by two years.

With roughly 11 million people dying per year, that's 22 million lives saved.​

That statistic will be at the front of my mind as I visit Bletchley today.

Enigma was not just one type of machine. It has a long history and went through many iterations. Click here to view Enigma's timeline or you can play with your own Enigma machine using this emulator.

The Bletchley Park website is a must - and I urge you to visit the physical site near Milton Keynes in the UK. Also, check out Dr Sue Black's book Saving Bletchley Park which tells you everything you need to know about the vital code breaking work done in WWII and the recent campaign to save the site.

If you want to find out more about encryption machines - then check out the Crypto Machine Museum (which is all online and pretty awesome).

And here's a great video dedicated to the inner workings of Enigma from the Perimeter Institute:

Last week, I went to see one leg of Professor Brian Cox's UK tour. I wasn't sure what to expect and, with two degrees in physics and working as a freelance science writer, I'd even say I was a little arrogant about the show. Was I going to learn anything new?

Yes, but not in the way I expected.

The lecture itself was a tour de force - covering everything from cosmology to evolution. Equations were put up and explained. It was engaging, it was fascinating and Brian struck the balance between making science simple - but never dumbing it down at any point. 

A Twitter-based Q&A session was also held with Brian's self-proclaimed sidekick and fellow Infinite Monkey Cager, Robin Ince.

My favourite question from the Q&A has to be "what does space smell like"? Brian postulated that as 80% of the matter in the universe is dark matter and that doesn't really interact with anything (let alone smell), then space doesn't smell of anything, really.

Another question struck a chord and has given me a whole new perspective on my work. Brian and Robin fielded the usual "Did the moon landings really happen?" question - but there was another part to this question:

Hmm. Well, we can't unless we go up to the Moon and see if the footprints from Neil Armstrong and Buzz Aldrin are actually there. Even then, I'm sure someone will come up with a far-flung conspiracy theory.

So, what's the answer?

Brian went off on a wonderful tangent about trust. About how we have to trust what we are told by scientists. We may not understand the nuts and bolts of the scientific theories and work done behind the world's laboratory doors - but we have to trust what scientists are telling us.

And that's a tricky point.

This question of trust has been rattling around my head since seeing the show.

I'd like to put forward my own answer.

We need to build trust by demonstrating that science is simple.

Science is not the pursuit of an intellectually elite few. Brian pointed out how, as children, we are natural scientists and the simplest questions that children have no embarrassment about asking, can sometimes probe some of science's most fundamental questions.

Science is simple. Yet, sometimes the methods and the language we use to teach and describe science make it feel utterly inaccessible to anyone without a PhD.

That's when the trust breaks down. And that's what we need to change both within the scientific community and outside of it.

Non-scientists need to keep asking questions about the world we live in. Scientists need to find a way to answer these questions without putting people off their work (check out my Sunday Science posts for some jargon-free science explanations).

We need to keep our children asking questions about the world we live in. Experiments should be at the forefront of any scientific syllabus - and exams should take a back seat. Science is not based on regurgitating facts memorised in textbooks.

It's based on theory and experimentation. It's based on applying scientific principles to test something new or challenge a pre-existing theory.

We are all capable of asking the important questions about the world we live in. And we are all capable of listening to and challenging the answers we get back.

That's exactly what Professor Brian Cox is doing with his world-breaking UK tour on science. 

He's demonstrating how science is big, complex but, fundamentally, it is simple and accessible to all.

We're all scientists. Science is simple. And we all need to leave our assumptions and arrogance about science at the door.

Myself included.

You can only imagine the magnitude of the geek shriek I gave off when Brian retweeted my blog post. Not only because he has 2.5 million followers, but because he called me his "scientific peer". That's totally going on my CV...

Is it a bird? Is it a plane?

No. It's a photon.

A photon is a fundamental particle of light.

But, just like Superman, it's difficult to pin down exactly what the humble photon is.

This is actually the subject of one of the most important arguments in physics. For a long time, scientists could not decide if the photon was a wave or a particle.

And, here's the really weird part, they discovered that the photon is both a particle and a wave.

It's like Superman and his alter ego Clark Kent have merged into one entity. And it took the superheroes of science to resolve this issue - including Albert Einstein and Max Planck.

Light has a split personality. It behaves like a wave and a particle.

The wave properties of light are easy to observe. Light reflects and refracts, for example.

But when you shine light onto a metallic surface and electrons are released (a phenomenon called the photoelectric effect) then light does not behave like a wave.

Building on Max Planck's black body radiation theory, Einstein explained the photoelectric effect by proposing that light is localised into small bundles - which were later called photons.

In other words, photons act as both a wave and a particle all of the time. This is known as the wave-particle duality. 

In the simplest terms, because a photon has no mass.

Particles gain mass as they travel through something called the Higgs field. Different particles interact with the Higgs field with different strengths to acquire different masses.

A photon does not interact with the Higgs field, so it is massless and can travel at the speed of light.

It's also incredibly difficult to capture a photon on camera to see what it looks like. Some experiments have come close - but, whether it's a bird or a plane, the photon looks unlikely to ever reveal its true shape.

If you want to get your teeth into some serious science, this paper will tell you everything you need to know about photons. And here's a more in-depth look into the photoelectric effect.

And this video explains to wave-particle duality brilliantly: