AI News, Music and Artificial Intelligence. Please don't shoot the piano player ... artificial intelligence
Gandhi said, "An eye for an eye only makes the whole world blind."
is truly amazing in balancing linear storytelling and horrific acts of violence, demonstrating the impact of the aftermath.
seen through the eyes of Eric Bana's Avner is a powerful allegory that even in the most just and noble fights against terror we eventually become that which we despise.
Mossad team leader Avner played by Eric Bana is absolutely riveting as the man who begins this righteous cause only to find that the cost is his soul.
Anver though an inexperienced operative and not an assassin is selected for the covert mission by Ephriam (the great Geoffrey Rush) for being a strong and effective leader of men.
they technically don't exist.In accepting the lead, Avner must leave his beautiful and pregnant wife Daphna (a very strong Ayelet Zorer) for what could be a number of years.
Avner pays large sums of money for information on the whereabouts of his targets from Louis (wonderfully shady Mathiew Amalric) and his wealthy Papa (weary and noble Michael Lonsdale).
Avner soon finds that whomever he kills is eventually replaced, and that he and potentially his family is now a target for the terrorists he was assigned to hunt down and kill.
The 2020 A to Z of Innovation Jargon
After working in full-time innovation teams for many years, I’m fully immersed in the culture of “Innovation”, with all that entails — agile methodology, ideation, guerrilla testing, arguing and more.
For anyone out there new to the wonderful world of innovation this guide should help you navigate your way through the transformational jungle of buzzwords, covering everything from design methodologies to hardware to cloud vendors and more.
As innovation is about pushing past preconceived boundaries and ideas, and it’d be good to get a definition of what innovation actually means, I’m going to start with: Defining Innovation is a bit like asking people from Ireland* which bar serves the nicest pint of Guinness;
Making money from new products is difficult, saving money by automating old processes is easier, so that’s what gets called innovation.
Ok now that we’ve cleared that up, let’s kick off properly: As in agile methodology, the opposite of old school waterfall style project management.
Regardless of whether you’ve used it before or not, it’s essential for innovation, where short, impactful projects are preferred to year long monsters.
More info here: https://www.agilealliance.org/agile101/12-principles-behind-the-agile-manifesto/ AWS is Amazon’s pretty incredible ‘cloud computing platform’, but that’s a bit like calling Amazon itself ‘somewhere to buys things online’.
AWS is a an ever expanding suite of technology products and services that has and is continuing to revolutionise both how companies deliver their technology solutions and also the businesses themselves that they support.
As an engineer, AWS makes my life much easier and is abstracting away a lot of the menial technical tasks that I had to do before — like looking after servers.
I didn’t, partly as the concept of what is basically a pilgrimage to worship at a tech alter makes me slightly uneasy for a number of reasons.
It can be quite entertaining watching your four year old scream “Belexa, play Little Mix!” in frustration at it, but as to whether your company should create a “Skill” (app) for it, really depends on the use-case and whether you think your customers would engage with you via voice assistants.
The “democratization” of machine learning, via cloud services, has opened up this field to engineers, not just data scientists, so expect much more projects in 2020 to include/focus on this.
To steal a line from one of my previous posts, if the person telling you that blockchain is the answer is wearing a lovely suit and is a “Senior Manager of Blockchain Technologies” in An Important Consulting Company, take a leaf out of Douglas Adams’ book and ask them what the question is.
Worth mentioning here as Tesla actually takes risks on new products … AKA real innovation ;-) I may not ever own one as I wouldn’t fit my 4 kids, one cat and one dog into it, but I’d definitely like one.
if you have a group working on a problem, everyone comes up with their own 8 solution ideas — this generates more ideas but also gives more introverted group members a chance to be noticed.
More info on their journey here, and official info on Deepracer here: https://aws.amazon.com/deepracer/ Not happy with mini racing cars, AWS also just launched the Deep Composer, a musical keyboard to, again, help people learn machine learning.
A colleague of mine recently showed me how he used photos taken from his drone to create a 3D model of his house, which is pretty cool for something not too expensive bought online.
If you can actually disrupt a common way of doing something, and make money from it, you’re probably going to end up with private jets and a house somewhere nice in France, but it’s really difficult to achieve.
It’s commonly quotes in tech articles that truck drivers are going to lose their jobs en masse, but I’d expect taxi and bus drivers to go first.
More info here: https://ethereum.org/ Event storming is another type of agile workshop that is used to quickly explore business domains / problems.
Not joking this time — several companies are exploring “flying cars” which are basically large drones capable of transporting people comfortably and clearly.
5G is the latest cell network that promises vastly superior internet speeds, which (should) lead to much more data-heavy applications being possible on mobile phones.
More info here: https://www.microsoft.com/en-us/hololens Horizons 1, 2 and 3 are a popular method of defining work / growth strategy in terms of how short, medium or long term they are.
Ideo is a famous design house, more info here: https://www.ideo.com/ IoT is another buzzword that consultants will tell you will be worth $50 billion dollars by 2025, but are never too clear about when the $50 billion dollars will actually start.
This video is from an iPhone project I made using more or less out-of-the-box functionality in iOS 13 when it came out: Just because the phone itself doesn’t seem that different, doesn’t mean what it’s capable of isn’t evolving rapidly each year.
More info here: https://www.bbc.com/news/science-environment-50267017 Nreal are a China based start up working on AR glasses, and are being sued currently by Magic Leap for allegedly stealing designs and tech!
https://www.nreal.ai/ No, not a joke about a 1970’s technology, pretty much every business team I’ve talked to over the last year has mentioned two things — documents and using OCR to get meaning from them or automating their processing.
Another programming language, I’m mentioning it as it’s both a great way to introduce people to coding and to do really complex coding for machine learning!
Basically a way of carrying out certain calculations that previously would be impossible / take too long to run, it can store information in a spectrum of values as oppose to the usual 0 or 1 i.e.
More info here: https://www.spacex.com/ No self respecting innovator is complete without a quiver full of Steve Jobs quotes, ready to impress people with at a moment’s notice.
(basically what I say to my boss anytime he tells me to do anything) It’s really clear that the most precious resource we all have is time An incredibly simple but important quote — time is the main thing we’re running out of, we need to focus on what’s important.
The product sensibility and the product genius that brought them to that monopolistic position gets rotted out by people running these companies that have no conception of a good product versus a bad product.
“With graphene’s amazing tensile strength, space elevators may be best way to get payloads into space, at cost effective prices”.
Good design that’s hard to improve on and does only one job but does it better than anything else — can you think of anything better at keeping rain off, other than umbrellas?
Vantablack is a super-black material, which it itself has been superseded by another, even blacker, material: http://news.mit.edu/2019/blackest-black-material-cnt-0913 Of pretty much no business value that I can think of, but I wanted another V, and who knows, maybe one day the CEO will be desperate and you’ll raise a trembling hand and say,”Well … if it really needs to be dark …” Ok, I’m struggling now, but I’ve 3 daughters and we’ve watched the new Wonder Woman 1984 trailer about 60 times together.
Nearly every TV series, cartoon, or movie has a male character as the main protagonist, so a female-led superhero film gets my vote for innovation.
Is AI the Future of Social Media Marketing? A Conversation with Cortex’s CEO, Brennan White
She actually went to completely different…she graduated from high school in a different town like, basically, end of my elementary school, she goes off to college, we moved to Western so actually, we went to a different town.
And that even includes my extended family I've got one cousin, couple aunts and uncles not too much in terms of the numbers on the family, but very, very cool family, mostly based in New Orleans, like I mentioned a moment ago.
We get somebody delicious food were you know Cajun background so that food is very delicious to us and I am looking for an opportunity you know Boston's not really known for that spicy food so it’s a nice break.
Shane: Now maybe the Cajun food like you can't tell me enough about that like any kind of like Creole or anything like that it's like over the top I'm a huge… the food I is just tasty food like I want something when I taste I'm like damn that's awesome I like that so I like my beer my wine like something that really kind of pack some punch to it so that's yeah I could only imagine the food that you guys are having out there for your family reunion, maybe I'll make it out there one day once I become part of the family or something like that you know who knows?
There's a lot of incongruous background like I was a music major in college a lot of my professional contacts don't know I have that background that's usually surprising, I also played rugby for eight seasons and was captain of a rugby team and played in England and Australia and stuff so there's a few of those but nothing too crazy, nothing like super wild.
And actually when I went to school in Chico, California I actually went to a number of schools but Chico where one of the places I ended up opening a bar, but I was actually going to start rugby I really love the idea of rugby, of course, it's a learning curve, right?
My family was like, California hippies so they're like oh let's hug it out instead of like smash each other and I'm like I kind of want to smash a little bit and so anyways I didn't get into that stuff till later on in life but huge fan of rugby man, it's just a camaraderie and like the way they treat each other and all that kind of stuff is just an awesome sport.
In my experience, there's like a secret business rugby network that works really well where you meet someone who played rugby you know they've got a trophy in the corner of their office with a like a ball leaning against the wall or something.
And you connect over that you end up being really good relationship because A, you know you've got the connection and B, I think rugby itself actually lends itself to good leadership if you played a leadership role in the team because it's like football and that you've got plays and you got to get 15 guys all kind of doing the same thing.
But it's live you know, never stops or almost never stops until you don't get… you know, you don't have the benefit of football where it's kind of from the same exact position, you're gonna have to talk and communicate on the fly.
Shane: Yeah Yeah, one of my best teammates over the years was a Jesuit guy in high school and played rugby there and went on to actually be in the Super League, which is kind of at the time was the best league the US has now we've got a major league professional rugby league that just started a couple years ago, but prior to that, he won the Super League, super young guy and was a Jesuit guy, so I'm familiar with that team.
Shane: That's awesome, man talk about a close knit community, that is one of the things too on top of the networking so my wife's brother Don, his name's Don Padlock, but he I'm telling you the community he goes into any area and they start talking about rugby and then all of a sudden they're like, whatever you need anything it is a nice little… the community is tight there which is awesome.
Brennan: Close to New York City, Yeah, you can just hop on the train as a college kid and be in the city and like an hour, hour and a half it's pretty convenient for partying and seeing the museums and going out with friends and all that good stuff.
But yeah, I mean to me as a growing up in western, as I mentioned, we had one of the best public schools in the state, but it didn't have a music department in terms of, there were no music theory classes, you couldn't really learn about music other than, being in choir or band or something like that.
But when I got to Vassar, I took some kind of intro music series courses and I just said, Oh, my gosh, there's this whole language behind this thing, that I really like it, you know, this appeals to like, I've got the creative style, and I've got the kind of math and nerdy side and music is kind of like the meeting of those two minds, right?
So I took on three jobs right out of college and just lived at home and save money, the main job being a, I was a software sales guy for tech company in Boston.
You grind it like I do you like so during the week, I was doing this, at night… because I used to own a bar I owned a bar in Chico I used to own a bar and some obviously real familiar… yeah the bartender side of things was always and then Chico was one of the number one, top 10 party schools in the nation at this time like playboy had mentioned it.
But that's a very large sales cycle, high relationship type business, whereas the SAAS business “Cortex”, you can actually, it's very analogous to what we did back then, which is, you know, lots of people coming inbound, trying to get them as much information as possible to make their decision as quickly as possible and as competently as possible.
And so, a lot of the stuff we learned, I learned like 12-15 years ago, is now you know, coming to super relevant stuff, you don't really know what the time is gonna end up being really relevant actually ends up being quite relevant.
Shane: Yeah, it's funny, I love that like when you talk about the foundation, like how people started off because I always think that's so interesting, like I used to do some door to door sales and I also worked for a real estate company.
So when was the last time you guys refinance because you know what the current… You know, I'm going this big old thing and actually you're turning people around and I'm high fiving people you know.
But I certainly didn't say that, I think every job, every role, almost every role, especially in corporate America, even if you're not, if you don't go with sales in your title, I think sales, especially an entry level sales role is definitely helpful for exactly what you were talking about.
Just getting used to talking to tons of people, you know, quickly assessing their personality types and whether they like, you know, direct information or they want to chat about stuff and just kind of understanding how to communicate with various people.
In fact, actually the impetus to start my agency, my first company was the guy in the cubicle next to me, I was 23 he was 43 he was divorced and had alimony and everything… Shane: Ahh motivation Brennan: Exactly and I said, you know what, 20 years from now I can be this guy if I don't get it going.
And he had some experiences that kind of echoed mine, where we said, hey, you know, we were both the class of 05 in college, which if you do the math, we were the last or the first class that was on Facebook in college, he was at a different school than me, but we both were in that initial 20 schools that could be on Facebook before the general public.
And then you know, as 23, 24 year olds, you say, hey, you know, businesses are going to want to use this and you know, you see people kind of taking stabs at YouTube or early Facebook before they were brand pages or right when Twitter launched and trying to figure that out and saying, they need a young person who actually uses this stuff to translate for them.
And that's actually where the agency came from and the Pandemic Labs was specifically to help people use social, in fact, as far as we can tell, where the first Social Media Marketing Agency in the US, and so we were kind of a little too early, you know, we started in early 2007.
And when the market did turn around, we'd already had some deals with Puma and Dunkin Donuts and stuff and so we were going to be the only game in town, and that led to, big booming business for creating content for ads, and social and all that stuff.
So, if you can imagine we're working with the big Fortune 500 brands kind of globally, and we're seeing the same exact problem at every company, at every level, which is that whether we're talking TV ads or online video or social, or any of that stuff, as long as there are visuals involved, there's an entire guesswork, you know, the entire thing is, is predicated on somebody guessing a piece of content and doing the best effort to create something cool that they hope the audience will like.
And then some CMO somewhere, kind of rubber stamping and saying, I don't know what's going to work, but I gotta try something so, for me, from my perspective as the agency that's kind of, we were on the hook for the success of all this, it is really hard to repeatedly be successful.
The kind of impetus came out of trying to solve that underlying problem, with a lot of ad tech out there that does a lot of kinds of specific ends of the funnel things but the thing we were trying to deal with was, how do you predict the performance of these visuals, whether it's an image whether it's a video because if you could do that, then you can invest with confidence across all these different channels.
We were both business guys and so we connected with some guys in Boston very quickly that happened to be machine learning guys that said Oh, yeah, this is a perfect machine learning problem we can do this, then we were off to the races and as you said earlier, that's basically like, four and a half years ago.
And same thing with Cortex we look at this and we go, okay, like, what you're talking about is saying, hey, let's predict what piece of content is going to be going to produce the right types of outcomes, right, which is kind of, that's the hardest part of marketing, right?
If we solved the question mark, all of a sudden we'll know we have a much, much higher level of confidence in what we're doing across any of the channels if we can figure out what the content is going to do in advance.
I want to know, I'm going to produce 10 pieces of content, which ones are going to have the higher likelihood for being successful because then I don't hopefully, have to do the other two or three or four and at least I know what kind of content is going to resonate the best.
Brennan: Yeah, and especially because I think a lot of people are high level thinkers, which I think is actually good and I think they think humans are creative, and I haven't seen a computer or software be creative.
But actually AI in the real world, the way it's taking over our lives today is via these super narrow capabilities so it's absolutely reasonable to say, humans are the only things capable of doing creativity today, that's super reasonable.
But even two years from now, it's absolutely reasonable to say, the totality that I think of as creativity is only doable by humans, but between now and then, little pieces of what you previously might have called creativity now, you're just going to say, oh, software does that.
In our case, it's looking at millions of images, processing it all with machine vision, knowing exactly what colors are being used, what objects are in the image, whether they're humans in the image, what age they are, what gender they are, where they're located, all that stuff, and using a different type of machine learning to find the patterns.
And the humans are getting much much better because the inputs into what they do, when you go to make an ad campaign, if you go into that creative process, knowing the colors you should be using, what the audience likes to see, what exactly, what kind of humans should be in it, to make the audience happy where the ad campaigns take place, all those kinds of things that just makes your creative process much more successful.
But these are things that we're going to make easier for you to be able to put together the pieces of the puzzle, for you to make that decision as a human, let's say I think the way you explained it, I think, it’s unpalatable right because now you go, Hey, we have these machines and we're going to look at a few colors and we're going to tell you which one works people go, well that sounds freaky, well what about like, I mean, am I involved, am I not involved?
Like it's just going to make things easier right we're here to take a look at things that maybe you might process with your brain you look at you go I think they like this, like this but we as humans, I can't take in 10,000 pieces of data, process it and then say hey, this is what we're getting and that's where the deficiencies happen.
And it showed that this kind of specific type of thinking actually gives you the exact hex code, right because we're not dealing with humans we're dealing with software with the ability to be super duper accurate.
And then later in the kind of findings, it had a slightly darker shade of pink, and that hex code for that and it said that this pink will detract, put up the numbers, like 18, or 17 is similar number, but the negative side from performance and they said, wait a second, so just by changing our images from this darker shade of pink, to this lighter shade of pink, we captured the data for 35 plus percent in performance?
And the cool part was they shared with us that they tried to do a color analysis before and because humans can't really see at the pixel level and can't really see necessarily reliably between, you know, the light pink in the dark pink, and they wouldn't know exactly where to draw the line anywhere, when they were doing this campaign, they did effectively, when they did their own analysis, they did like the 12 colors that you get in the crayon box.
Or are they more valuable to the company now, right now that they know they can make those changes across all their designers, across their photographers, across their campaigns and all of a sudden you know, the company performs better simply because the ad can cost them… and that's just an example kind of one change of course, it's looking at thousands of these things and communicating them all very quickly and easily.
And yeah, back to that story you're kind of getting at that's the crux right is that in every case where automation you know, screw AI, but in every case where automation has appeared, the worker has become more valuable to the company not less.
When we went from manually assembling Ford Model Ts on an assembly line, one person doing each screw, there's like, you know, five employees for the whole car assembly, because a lot of it is automated, those humans are responsible for much more value delivered than the thousand humans it took before.
Now you want to optimize the use, you want to guarantee that kind of data is everywhere that it needs to be that we're taking maximum value from this data, we have a platform that is fully operational SAAS platform that not only tells you the data, but also automates a lot of decisions for you, right?
Because if you can think about this, where this is going, if Cortex knows the colors that your audience likes, and it knows what should be in the images, and it knows all these creative choices, and it knows them by platform, and by the time of day, and all these things, because it's using clustering, to compare all of those inputs to find patterns.
So like, you know, for example, if Google put out tomorrow, some crazy image processing capabilities that we don't have now, like, you know, you can look at an image and look at the angle of the sun and figure out the exact time it was taken and making this up.
But let's say they could do that they're like, we could just immediately deploy that to make our data set more accurate and better, that's effectively what's happening every day, somebody rolls out this new thing that we can just bolt on to the processing part at the beginning of our process when we gather the data, and process it.
Shane: It might be sooner than later be careful I'm a big fan of Boston I was long story short actually, I was supposed to I was either gonna move to Boston or Chico like after I saw Good Will Hunting which is I talked about this on another podcast but love that movie Good Will Hunting.
So tell me when we're talk a little bit about a content are there any companies you look at like they're creating Epic Content, or maybe even a client that you guys are maybe tie into that because I don't want to say, just look at something and say they're creating Epic Content, or is there any evaluations you guys have done and said this company is like, they seem to know what's going on when it comes to certain types of content?
Brennan: Yeah, of course, yeah so take one step back first you know, if you think about what we're talking about, you know, we've kind of talked about it from the brand marketer perspective, but even put the other hat on.
Yes, there are several examples, kind of on a spectrum, by far and away the most committed to that vision is Red Bull, as I'm sure we're familiar with, right, they're a drink company, but they have an entire division that is self-sufficient, financially, that is literally just running cool events that their audience would like to look at, exactly they put a guy into space so he could jump out of space and parachute down.
They do all those weird you know, like boats jumping off of platforms into water, things like all those events, but they completely committed to the idea that giving the consumer something that the consumer values is the point right and then they kind of literally their company is divided into the drink apart and the content part so they've completely committed to this and as you can see that they're never going to run out of stuff to talk about they're never going to not be you know constantly in the media.
Nowadays they probably are more sophisticated I don't know if they're using AI like we are and if they aren't they will just call us but I mean, yeah, ultimately, like they are committed to it with both feet and… Shane: You know, another thing is about Red Bull and I read a book on this and they talked about Red Bull this was interesting because it wasn't maybe 10 years ago, whenever they started up.
And so I started reading this book and they talked about what they did with Red Bull would do is the psychology of Red Bull obviously is you're getting some kind of you know, endorphin rush, right, you get something where you're drinking it, you're kind of this heightened sense.
So they would hand these things out so whatever, 10,000 people are drinking Red Bull or whatever the number is 1000 watching their favorite artists having the time of their life, there with their girlfriend and they're thinking about Red Bull, so every time they will go to a market or someplace they see a Red Bull they think about the experience.
Now, there's so much content, the average brands getting 15,000 pieces of content per year, just to be competitive, on social and other different things all wins and if you rewind 25 years, right back in the let's say the 80s right before the internet entirely, they were creating what a few dozen commercials a year a few dozen ad spots and magazines, a few dozen print.
Red Bull just got in front of it and leading the way and you know companies even like Tesla, they're not set up like Red Bull is, they haven't jumped in, in that regard but they know as you can tell like they know that the best thing they can do, the thing their audience wants is tidbits of videos or content about these cool new futuristic products.
So even though they're very product focused and they haven't done what Red Bull did, they didn’t build a giant division they spend almost nothing on marketing Tesla, but they're actually philosophically very similar and they lead with their story and they lead with you know, we are building the future and here's a glimpse at all this cool futuristic stuff and I think you know, even though in terms of their spend there can be the exact opposite of Red Bull, I'd still put them in that category of the company, you totally get that.
And the amount of orders they put in like just recently the one they did with the cyber truck and that there was a company, broke the window and it sounded like Yeah, but the end of the day they got millions and billions of dollars in sales, it's like the things that he brings out and have a price point and the amount of thought they put into it, it's one of those like, I look at these and I go why hasn’t anybody else come up with this like it's so weird to me.
In fact, during their presentation that you saw, they took the logo off of all the major selling pickup trucks and you really get down the part they all look basically identical functionally the same, in fact that you know, they took a stance here aesthetically, right?
They’re certainly not going to appeal to the masses, but they don't need to appeal to the masses right to be one of the biggest selling trucks of the year they sell 100,000, 200,000 of those and they're off to the races right.
We do go into every single presentation, having just process a bunch of data, either about companies we know will be in the audience or industries that you know if it's an industry conference, or whatever, the one I was at Thursday was in Belgium, and I went in there and we processed the entire beer industry, because Belgium of course are very passionate about their beer and they do beer quite well.
And then we looked at one company, which was Deloitte, and we just looked at their built in office and we were like, this is what your Belgian office should change to target their marketing more and we've kind of went through that as the presentation and it's super visual, it gives them immediate takeaways and of course, we can do it at the push of a button.
So super easy, just you know, the day before two days before, to kind of get the companies from the guy running the event and say, you know, who's gonna be in the audience who should we tell this to, process all of that and share it and so Shane: I love it.
I think Elon Musk actually it not sound like a fanboy this entire thing, but I think if you think about it, if my goal is to kind of level up my own software, I think you spend dinner with him, and you just kind of because not to compare ourselves in the slightest, but he attacked every problem at its roots.
And I kind of want to know how you know like, what is his process that kind of way the urgency versus the you know, the importance versus whatever and how to make tough calls you know, in some of the toughest scenarios out there you know, during the Cold War and World War Two all that kind of stuff.
- On 26. februar 2021
These Google AI experiments are crazy! This is the future.
Google has some insane experimental AI - I figured it was worth exploring to see how close it is to destroying us all. Check them out yourself: ...
Mr Shadow: a song composed by Artificial Intelligence
Scientists at SONY CSL Research Laboratory have created the first-ever entire songs composed by Artificial Intelligence: "Daddy's Car" and "Mister Shadow".
Sophia the Robot and Jimmy Sing a Duet of "Say Something"
Jimmy demos amazing new robots from all over the world, including MIT's Mini Cheetah, a wearable tomato-feeding robot and Sophia, who returns to the show ...
Artificial intelligence plays GRADIUS (HUMAN LEVEL AI)
This video shows a robot playing gradius. In the game, he is doing primarily three tasks simultaneously: 1. navigating a plane. 2. shooting enemies. 3. managing ...
MAKE A SONG WITH YOUR FACE | A.I. Experiments
Have you ever wanted to make music with a robot with your face!?? The future is amazing! Googling Myself ▻ ..
The Flaming Lips collaboration with Google AI bears fruit at I/O 2019
Google AI's Magenta team and The Flaming Lips teamed up to create an AI-assisted performance as part of the headline concert at I/O 2019. The result was Fruit ...