Where do we stand?
Diversity, AI & Education ft. Gigster
Many have spoken out against Trump since he proposed banning Muslims from entering the United States. This includes Sundar Pichai, CEO of Google. In his first post [2min]on Medium, he shared about his experience as an immigrant coming from India. He also points out that his experience is not unique to himself. While he did hesitate to publish the post, but luckily he did.
I feel we must speak out — particularly those of us who are not under attack. — Sundar Pichai, CEO of Google
Other than him, other leaders have also shared their views on the topic, notably Mark Zuckerberg, CEO of Facebook’s post [1min]. He assured that Facebook will protect rights of Muslims in the community and are welcomed there.
In Obama’s terrorism speech [13min] two weeks ago about keeping the nation safe. He not only listed out the actions that will be taken, but also further stated things that should not be done. The latter list includes not going to another war which will lead to sending another generation of young Americans to fight and die for another decade on foreign soil. Other than that, he also points out that groups like ISIL are only a tiny fraction of the Muslim community globally and does not speak for Islam.
We cannot turn against one another by letting this fight be defined as a war between America and Islam.
On another controversial topic, Artificial Intelligence (AI), a group of tech luminaries including Elon Musk, Reid Hoffman, Peter Thiel, Jessica Livingston and Sam Altman have taken a stance by pouring over a billion into a newly founded non-profit venture, OpenAI [3min].
In an interview [11min] with co-chairs of OpenAI, Elon Musk, CEO of Tesla & SpaceX and Sam Altman, CEO of Y Combinator, Steven Levy gave us an insight about the formation of the venture and where it is going.
OpenAI is a venture that will be freely owned by the world, as it has no obligation to maximize profitability. It is also uncertain that companies like Google, Facebook and Microsoft which shared a lot of their research in the field now, but Altman thinks that there is still question of how much they will in time to come.
All OpenAI’s research will be open sourced and usable by everyone. So there is a high probability that OpenAI’s research will be used by Dr. Evil, Altman thinks that its far more likely that many, many AIs, will work to stop the occasional bad actors. Else, we’re really in a bad place.
I think the best defense against the misuse of AI is to empower as many people as possible to have AI. If everyone has AI powers, then there’s not any one person or a small set of individuals who can have AI superpower. — Elon Musk, CEO of Tesla & SpaceX
Despite growing capabilities of AI, Deep Learning and Machine Learning, we humans actually still do not know what goes on in the machine when they are making the decisions. It is like a huge black box, which is unlikely for a mathematical technique. In this article [4min], Ingrid Daubechies, former president of the International Mathematical Union explained how a typical supervised-learning system would work. She concludes that despite these methods are already leading to interesting and useful results, but many more techniques will be needed.
Google’s Deep Dream [22min] is one of the efforts that has proven useful in doing just that, giving us a glimpse of how the machine recognise objects. This project reconstructs items that the machine perceived in an image based on its knowledge of objects. Some people also thinks that when this algorithm is applied to images, it has similarities with hallucinations when we’re on drugs.
One of the notable findings (figure below) in this study [5min] shows that the machine perceives dumbbell with the hand instead of just solely the dumbbell. This might make us question if a dumbbell is really a dumbbell without a hand holding it.
To some, AI maybe too far out from their league. In celebration of Computer Science Education Week that ended yesterday, almost 200,000 Hour of Code sessions have been organised. Hour of Code gives an hour of crash course into the basics of programming covering functions and loops, with its hope to demystify that coding is hard.
In one of the Hour of Code session hosted in an Apple Store in NYC, Tim Cook, CEO of Apple crashed the event [5min]. While he was there observing the session and gave a brief speech to the students, he also spoke to Lance Ulanoff of Mashable.
Cook and Apple really want 21st century classrooms to function less like a traditional lecture and more like the room we were standing in, with excitement, engagement and — especially — real-time collaboration.
I’m not a fan of teaching to the test, I think creativity is so important. Training the mind how to think is so important. Teaching to the test, to me, is too much about memorization. In a word where you’ve got all the information you’ve got right here, [gestured to an iPhone] your ability to memorize what year a war was won and all this kind of stuff isn’t very relevant. -Tim Cook, CEO of Apple
While Hour of Code exposes people to coding, Mike Hadlow doubts that these short courses will help to produce professional programmers. In his post [10min], he shared that even Computer Science students are not getting hired because they lack the competency to code, while most of those who code learnt were self-taught.
This brings up the question, “Should coding be on of the core subjects taught in school?” Despite rising popularity of online coding academies and coding bootcamps, I see coding being a skill that is increasing scarce given the complexity of task are increasing and there is more distinction between what a coder sees and what an user see. Think decades ago, where everyone uses Command Line Interface to communicate with computers (thus some degree of geekiness), while users now simply activate their devices with your voice by merely saying phrases like “Ok, Google”, or Siri or Cortana or Alexa. Thus users will only be more ignorant of what it takes to deliver such sophisticated application while coders will need to be more competent (thus higher barrier) to be able to deliver at such standards.
Earlier this week, Gigster raised $10M from notable investors include Andreessen Horowitz, Y Combinator’s Continuity Fund I, Ron Conway of SV Angel, Ashton Kutcher of Sound Ventures and Jason Calacanis of Launch Fund, as reported on TechCrunch.
Gigster turns your idea into a working product and you don’t have to code, nor hire. Just send them your app idea and you will get your app. While Gigster is still using freelancers to develop the app for users, and only using it’s AI engine to convert a client’s product proposal into a development plan, and helps Gigster’s army of remote developers plug in pre-made code blocks to efficiently build the app.I don’t see why there will not be room an AI to generate another software thus replacing the freelancers in the equation just like Uber’s plan [1min] to replace their drivers with driverless cars. It might not replace every single developer on earth, but it will definitely help to reduce demand for developers and to move developer’s focus on more challenging tasks that the AI can’t solve, yet.
[Resuming my earlier point on not everyone needs to learn code] Maybe, this is another reason why it is not that important after all to teach coding as a core subject in school.
Lastly, another one minute read to founders who need their asses kicked.
