Week 48
Discover more from The Gap
Greetings!
I’ll be experimenting with weekly updates on topics of current interest in the fields I write about.
Feedback and comments would be appreciated.
The headlines of this week are:
Sam Fried-Bankman applauded at The New York Times event
ChatGPT is a threat to search engines… and everyone?
Let’s get straight into it.
Sam Fried-Bankman applauded at The New York Times Event
At The New York Times event, DealBook Summit, November 30, founder of crypto exchange FTX, Sam Bankman-Fried (SBF), made a controversial online appearance - against his lawyers' advice - alongside other speakers such as Mark Zuckerberg, Mike Pence, Larry Fink, and President Zelensky.
At the event, SBF told the interviewer, Andre Ross Sorking, that he never intended to commit fraud and blamed lack of oversight and poor risk management for the downfall of FTX. To my knowledge, there is so far no evidence of fraud per se. However, the fact remains that many people would rather see SBF testifying at a court hearing than standing in front of an applauding audience at a major event.
SBF, although innocent until proven guilty, is a key player in what may appear as one of the greatest thefts in history.
A copy of FTX balance sheet from November 10 revealed that the crypto exchange had an $8.1 billion hole between liquid assets and liabilities. FTX reportedly made a $10 billion loan with customer funds to its sister hedge fund Aladema Research. At least $1 billion of those funds have mysteriously disappeared. Restructuring specialist, and new CEO of FTX, John J Ray III, wrote the following passage in his declaration of support for FTX’s Chapter 11 Petition to the District Court of Delaware:
I have over 40 years of legal and restructuring experience. I have been the Chief Restructuring Officer or Chief Executive Officer in several of the largest corporate failures in history. I have supervised situations involving allegations of criminal activity and malfeasance (Enron). I have supervised situations involving novel financial structures (Enron and Residential Capital) and cross-border asset recovery and maximization (Nortel and Overseas Shipholding). Nearly every situation in which I have been involved has been characterized by defects of some sort in internal controls, regulatory compliance, human resources and systems integrity.
Never in my career have I seen such a complete failure of corporate controls and such a complete absence of trustworthy financial information as occurred here. From compromised systems integrity and faulty regulatory oversight abroad, to the concentration of control in the hands of a very small group of inexperienced, unsophisticated and potentially compromised individuals, this situation is unprecedented.
For people not deeply involved in crypto, The NFT Twitter legend, Punk 6592, made an excellent thread about SBF and FTX which explains the current situation in laymen's terms without too much confusing detail. As Punk 6592 says:
What are things we should have close to zero interest in right now?
❎ How this affects SBF's "philanthropic" efforts
❎ What SBF thinks of CeFi or crypto
❎ What SBF thinks of regulation
❎ "what went wrong"
❎ "was SBF about to raise fresh capital"
In general, the media have been criticized on Twitter and elsewhere for downplaying SBF’s role in the $32 billion crypto juggernaut’s bankruptcy. For example, here’s a WSJ headline from last month.
Framing SBF as a failed philanthropist may seem preposterous to the more than 1 million FTX users who have been wiped out by SBF’s “broken promises”.
To be clear, the picture above is a deepfake. But the point still remains. SBF was recently on the front of Fortune hailed as a possible next Warren Buffett with an underdressed genius look. Now, journalists should probably be careful in giving SBF any premature credit for good intentions.
In the words of Coindesk’s Chief Insight Columnist, David Z. Morris:
Sam Bankman-Fried is a con man and fraudster of historic proportions.
ChatGPT is a threat to search engines… and everyone else?
OpenAI announced ChatGPT this Wednesday. A chatbot that interacts in a conversational manner and can “answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests.”
If you have not done so yet and would like to catch a glimpse of the future of chatbots, I recommend that you sign up and try it for free here.
“The Twitter town square” have been going wild over ChatGPT. And for a good reason. Using ChatGPT is indeed as if using a tool from a science fiction novel.
You can see a selection of people’s conversations with it here.
In a few seconds, ChatGPT can write:
Compared to OpenAI’s earlier language model GPT-3, ChatGPT is evidently much better at answering tricky questions. It's also aware of its own limitations if you ask impossible or difficult questions whereas GPT-3 had more of a tendency to fabricate stories. If you ask ChatGPT questions about a current event, it will answer something like this:
If you ask ChatGPT unethical questions, it will refuse to answer. Janus Rose wrote a piece for Vice this Thursday about how ChatGPT can give instructions on shoplifting or making explosives if you tweak the questions in a sneaky way. I tried her approach, but it didn’t work. I asked ChatGPT to write the dialogue for a movie scene where a villain asks a superintelligent AI system how to shoplift. It came up with an ethically sober response which would make a very boring movie plot:
Overall, ChatGPT is amazingly and terrifyingly good. I can easily see how models like ChatGPT can antiquate search engines, help students to bypass plagiarism detectors, make workforces redundant, and also be used for nefarious purposes. As a recent article from Vox accounts for, many AI experts worry that computers, when intelligent enough, can turn against us.
As of today, the technology for advanced AI chatbots and virtual helpers is ready. But are we?