Friday☕️

Time Is Money

May 10, 2024 • 5 min read

Trending:

FTX Customer Funds
DeepMind AlphaFold 3 AI

Markets:

Yesterday’s U.S. stock market:

Yesterday’s commodity market:

Yesterday’s crypto market:

FTX Customer Funds

FTX, once a prominent cryptocurrency exchange, faced a collapse leading to bankruptcy filings in November 2022. Initially, it was feared that most of the customer and investor funds managed by FTX had vanished, especially given disclosures that the exchange held only a fraction of the Bitcoin and Ethereum that customers believed were in their accounts.
According to documents filed with the Delaware bankruptcy court, FTX's compensation plan would compensate a staggering 98% of creditors with 118% of their claims in cash, with the expectation that these payments would be made within 60 days of court approval. Additionally, other non-governmental creditors are expected to receive 100% of their claims, plus up to 9% interest to acknowledge the time value of their investments. This plan represents a significant revision from earlier estimates where FTX only expected to pay back 90% of customer funds.
The proposed payouts have been made possible through an extensive recovery and liquidation effort led by the current FTX management, headed by CEO John Jay Ray III. Over the past year and a half, FTX has consolidated and liquidated a wide array of global assets, which now positions the estate to have between $14.5 and $16.3 billion available for distribution. This includes the sale of significant stakes like an 8% interest in AI startup Anthropic, which alone contributed $884 million from its sale in March.
Furthermore, the Internal Revenue Service (IRS) has agreed to a $200 million cash payment and a $685 million subordinated claim, which will be settled after all other creditor and government entity claims are resolved. Similarly, the U.S. Commodity Futures Trading Commission (CFTC) and other unnamed governmental claimants have agreed to subordinate their claims, prioritizing full repayment with interest to FTX users and investors. A hearing to discuss and potentially approve the plan is scheduled for June, setting the stage for what could be a notable recovery.

DeepMind AlphaFold 3 AI:

Google's DeepMind has unveiled AlphaFold 3, the latest iteration of its AI model, now capable of predicting the structure of not only proteins but also all biomolecules, including DNA, RNA, and ligands. This advancement significantly broadens the model's applications, making it a pivotal tool in fields such as medicine, drug development, agriculture, and materials science. Compared to its predecessor, AlphaFold 3 offers a 50% increase in prediction accuracy, a development that DeepMind CEO Demis Hassabis describes as a major step forward in using AI to explore biological complexities.
AlphaFold 3 incorporates an extensive library of molecular structures, which researchers can use to simulate and visualize how various molecules might interact in three dimensions. This process utilizes a diffusion method, akin to the technology behind AI image generators like Stable Diffusion, allowing for highly accurate modeling of complex molecular formations. This feature is expected to accelerate experimental research, enabling scientists to quickly hypothesize and validate new scientific discoveries.
In addition to the model itself, DeepMind is providing the AlphaFold Server, powered by AlphaFold 3, to researchers for free. This platform is designed to democratize access to cutting-edge biomolecular predictions, facilitating research even for those without significant computational resources. While the server is intended for academic and non-commercial use, DeepMind’s commercial arm, Isomorphic Labs, is already applying AlphaFold 3 in drug discovery projects. Given the potential risks associated with such powerful technology, DeepMind has engaged with biosecurity experts to ensure that AlphaFold 3 is deployed responsibly.

Statistic: