#CNDTalks: Machine Learning, Investments, Fintech
As you probably know, we’ve recently launched #CNDTalks on our Twitter and Facebook. It’s a series of 12 discussions with specific questions about blockchain, crypto, and fintech. The conversations are picking up, thank you so much for your intellectual contributions!
Today we want to share with you some additional responses that we got. There are still several weeks to go and we invite you to jump in and share your insights!
What are the future implications of machine learning in everyday life?
We all hear a lot about AI and machine learning talking over every possible dimension of work and leisure. Yet our community agrees that it’s just the beginning.
Joris Debaere on Facebook noted:
The problem right now is computers can’t think for themselves, they are very good at performing given tasks. Humans however can think for themselves, that’s why we are (still) smarter than computers. Machine learning is part of the artificial intelligence boom we will probably see in the next few decennia.
On Reddit, cdin (Jonathan) outlined the vision for ML replacing some of our daily habits:
We are just beginning to see in what ways ML, neural networks, GAN’s etc. As we go on, we are going to see a dramatic increase in the ability of our everyday objects to predict and learn. It sounds obvious, but when you consider the long reaching and subtle effects I feel the shift will be dramatic. Right now every one of us spends hours a day searching google for things we want, yelp for where to eat — the reality is that none of this face time needs to be occurring. A personalized ML in our devices, probably synced to the cloud in order to follow you across devices/physical areas, will render all of that (and online advertising in traditional sense) obsolete. I predict in the next 20 years we have such a technology developed and deployed on a massive scale. Your personal ML will aggregate food and entertainment choices, dr visits and disease, exercise, stress and over time develop relevant, timely prediction offering you the best choices from those available not just for your situation, but your temperament at a specific time.
FC Crypto in our Telegram chat also looked at both sides of the issue:
And then there is another dark side of machine learning that is spoofing where humans might not notice change but machines can be tricked in to misclassifying objects.
But all said and done our future looks great by machines by our side.
We can expect a whole lot things right from automated public transportation to help in medical field, taking over challenging Jobs like in defense areas, solving climate change, virtual assistants becoming more interactive and responsive, helping in our daily chores at home, with access of big data and neural networks researchers will be able to make more accurate predictions in the financial market like CND.
How do you identify whether to invest in a project or not?
That’s a tough question especially in the current market and it caused a lot of debated on Twitter. Avoiding scams was a big theme.
@emiratekay1981 offered four simple steps:
1. Only invest if you are okay with losing all of your investment.
2. Look for teams with projects that already have a product or service available.
3. DYOR on the team and get on telegram to communicate with their community.
4. Believe in the project
MajorCrypto provided some additional hints:
First, I look at concept behind the token. How likely is it to produce an ever-increasing demand and storage of tokens in future? Next, look at team. Are they credible? Next, sponsorship. Is it relevant? Finally, number of tokens. Is it large enough for ecosystem but not too big?
cdin (Jonathan) presented an actual framework for reviewing a project, while emphasizing the need for regulation:
Crypto is so full of scammers, schemes and other ways to steal your coins. Especially ICOs. The people who are doing the scam exit ICOs have gotten very sophisticated. They know how to make a page that will pull on your strings and get you interested, they know how to fake an advisor page and partnerships. … The best thing you can do for due diligence is — Hang out. Go to the discord, ANN, and telegram channels and BE in the community. Get involved, get to know the team and see what they are about. At that point, you must rely on street smarts and judgement. If you watch a community for 3–6 mos you will start seeing patterns. The devs hide when there are problems, channels not well moderated, roadmap keeps changing, site changes, sudden white paper changes. All these things I am looking at. … I never thought I would say this in my life, but — we need regulation, now. The scammers and grifters are scaring away new money, including big money. I personally have stopped recommending crypto to the everyday person until this is handled somewhat. I wouldn’t be ok having to explain to my grandma why she lost 10k investing in an ICO. I really wish this would change as we need new people, and new money.
Risk management is still the key. FC Crypto highlighted the need to understand the risk:
Lower the risk by investing only as much as you are prepared to lose, don’t go all in on a project which doesn’t even have a ready product even if it does keep your investment to the minimum, let it grow with time and then invest some more when you find a good opportunity.
Be apart of their active community and discussion forums and clear your doubts if you have any.
Check their website and see if they are a registered company.
In many countries ICO’s are illegal, the best thing would be wait for government to regulate these ICO’s, this will in turn attract lot of genuine investors.
What is the next big thing in fintech?
Cdin (Jonathan) presented two big trend — decentralised exchanges and the move from SWIFT:
There are two developments in the fintech world I have my eye on at the moment. #1 (not by importance, necessarily) — UK eyes leaving SWIFT transaction system. That is a huge deal. The SWIFT system has been the underlying structure for bank based international payment and remittance since the 70’s. It covers US, Asia, UK… If UK breaks away from SWIFT and demonstrates a cheaper and more effective alternative, it will be huge and I would expect others to follow suit. It could be a huge boon for adoption of new technologies on an institutional level. and #2) DEX’s.
Decentralized exchanges are going to be huge 2018, with Binance slated to start their own. The central point of failure that is the centralized exchange system, along with its potential for abuse and manipulation is on the way out. And with that, hopefully a lot of the manufactured volatility used to strip investors of hard earned cash. It won’t happen overnight — liquidity will move in slowly from organic partners, as most DEXS have functionality built into their smart contracts to mitigate some of the effects of HFT, wash trading and other practices rife on normal exchanges. Holding your own assets and divesting them to smart contract, trading directly with other individuals is how this should have happened in the first place. I expect these principles for direct, trustless exchange between parties will bleed out into the rest of the fintech sector, hopefully bringing control to users and transparency to organizations.
Mortalwitness succinctly supported that:
The next big thing in fintech is smart contracts.
FC Crypto saw the potential for collaborations between banks and blockchain-based fintechs as as a significant development:
Financial institutions have realised the importance of technologies like Blockchain which will propel as well disrupt Banking services and therefore they are looking for ways to work with Blockchain-based fintech start-ups. Blockchain technology is changing and shaping the financial system. Some banks, companies, entrepreneurs and individuals have tested and are ready to embrace the technology. Thus partnerships between Blockchain-based fintech companies and banks, businesses is one of the next big things in fintech.
Fintech startups will provide banking services to unbanked, there are billions who are still unbanked but have access to their mobile phones and internet, these fintech companies use data sources such as web history, mobile phone data and social networking to determine the creditworthiness, this enables the unbanked to micro loans, even the small businesses would benefit with e-commerce lending and we will see lot of people adapting to this system.
Financial services are using Robo advisors driven by AI that can provide financial advice and give more personalised advice for decision making.
MajorCrypto on Twitter saw another big trend in automated advice:
I believe the next big thing in #fintech will be automated financial applications that advise, transact, and let you pay, lend, invest, and borrow money across financial institutions using #blockchain technology.
And what is your view?
Join the conversation, the next question is coming up on Twitter on Saturday, 1 September at 1 PM UTC.