SBF Appeals from Prison, 35-Page Document Accuses Judicial Misconduct

marsbit2026-02-11 tarihinde yayınlandı2026-02-11 tarihinde güncellendi

Özet

Sam Bankman-Fried (SBF), the founder of FTX currently imprisoned, is appealing his fraud conviction and 25-year sentence through a 35-page pro se motion filed by his mother. The motion cites newly discovered evidence and alleges multiple judicial and prosecutorial misconducts. Key claims include the absence of favorable witnesses like Ryan Salame, who was allegedly threatened by prosecutors, and coerced testimony from Nishad Singh. SBF also challenges the prosecution’s financial narrative, presenting a sworn statement from former FTX data head Daniel Chapsky arguing that Alameda’s account was misrepresented to show artificial losses. Additionally, SBF accuses Sullivan & Cromwell, FTX’s bankruptcy counsel, of undervaluing assets to support the narrative of insolvency, despite high customer recovery rates. The motion further suggests political targeting by the Biden administration and requests Judge Kaplan’s recusal due to alleged bias. Legal experts view the appeal as an uphill battle given the high bar for new evidence and judicial discretion in such motions.

Original Author: Sanqing, Foresight News

On February 10, according to Inner City Press, FTX founder Sam Bankman-Fried (SBF), currently serving his sentence at Terminal Island prison in California, is actively seeking to overturn his conviction. A pro se (self-represented) motion for a new trial, submitted on his behalf by his mother, Stanford Law Professor Barbara H. Fried, has been formally filed with the court. This 35-page document, citing Federal Rule of Criminal Procedure Rule 33 and newly discovered evidence, strongly requests the overturning of his 2023 fraud conviction and the 25-year prison sentence imposed in 2024.

The motion primarily argues that the trial was severely flawed due to the absence of key witnesses (such as former Alameda Research co-CEO Ryan Salame and former FTX.US executive Daniel Chapsky) from testifying; that prosecutors allegedly concealed evidence; and that the entire process was influenced by political factors, with SBF subtly suggesting he is a victim of a "targeted attack" by the Biden administration.

The evidence and arguments submitted by SBF this time are not aimed at directly proving his "innocence" but rather adopt a legal strategy questioning the procedural flaws of the judicial trial.

Core Accusation One: "Customized" Witnesses and Judicial Coercion

The motion accuses the prosecution of coercing and inducing key insiders to turn against SBF and silencing witnesses favorable to him.

For example, the absence of former Alameda Research co-CEO Ryan Salame. The motion cites Salame's public statements after August 2024 (including an interview with Tucker Carlson) as newly discovered evidence, alleging that prosecutors threatened to indict Salame's partner, Michelle Bond, to prevent Salame from testifying to SBF's innocence.

Regarding former engineering head Nishad Singh, who testified against SBF, the motion discloses that during pre-trial interviews, when Singh's initial statements did not meet the prosecution's expectations, a prosecutor angrily "slammed the table" and斥责 (chì zé - reprimanded/scolded) Singh's memory as "unreliable."

SBF believes that such high-pressure intimidation forced Singh to subsequently alter his testimony. The motion formally requests the court to order the prosecution to hand over the relevant interview notes to prove this coercion was concealed.

Core Accusation Two: The Disappearing "Liabilities" and the Mystery of [email protected]

SBF submitted a sworn declaration from former FTX Head of Data Science Daniel Chapsky, countering the misappropriation allegations from a data perspective.

The motion points out that the prosecution had presented the huge negative balance in the [email protected] account as ironclad evidence of SBF's misappropriation of customer funds. However, Chapsky's declaration refutes this, calling the prosecution's interpretation a "fundamental misrepresentation."

He stated that the negative balance in this account corresponded to cash and assets held offline by Alameda. The prosecution only showed the "debit" negative numbers to the jury but deliberately omitted the corresponding "credit" assets, thus fabricating a false impression of a multi-billion dollar shortfall out of thin air.

Chapsky's data analysis further shows that if correctly accounted for during most of 2022, Alameda's account on FTX actually maintained a positive balance of approximately $2 billion. The prosecution and expert witness Peter Easton deliberately displayed only certain specific sub-accounts with negative balances, misleading the jury.

Core Accusation Three: Bankruptcy Law Firm S&C's "Asset Erasure Technique"

SBF also targeted the law firm Sullivan & Cromwell (S&C), responsible for FTX's bankruptcy restructuring. He accuses S&C of artificially creating "insolvency" to align with the prosecution's conviction narrative and to earn exorbitant legal fees.

The motion points out that FTX held a venture portfolio valued at up to $8.4 billion at the time of bankruptcy (including investments in Claude AI developer Anthropic). However, early in the bankruptcy process, S&C and the prosecution, to solidify the capital shortfall, recorded these less liquid but highly valuable assets at zero or extremely low values.

SBF emphasizes that the bankruptcy team's eventual confirmation that customers will receive 119% to 143% cash recovery itself proves that his assertion during the trial—"FTX was solvent, the money wasn't lost"—was true.

Core Accusation Four: Political Targeting and Judicial Bias

Finally, SBF played the political and procedural cards. He implied he is a victim of a "political war" by the Biden administration. As a former major Democratic donor, he was quickly distanced from and harshly sentenced after the incident to quell public anger.

Furthermore, given that presiding Judge Lewis A. Kaplan repeatedly rejected defense evidence regarding "FTX's solvency" during the previous trial, SBF's motion not only requests a new trial but also explicitly requests Judge Kaplan to recuse himself, citing the judge's demonstrated extreme bias and inability to adjudicate the case fairly.

Is This Breakout Attempt Doomed to Be a Struggle of a Cornered Beast?

A Rule 33 motion requires the evidence to be "newly discovered" after the trial, which the defense could not have obtained through "due diligence" during the trial. The judge will likely rule that Salame and Chapsky were known potential witnesses during the trial, and the defense's failure to call them was a strategic choice or an objective difficulty, not "new evidence."

Moreover, FTX's high recovery rate (even exceeding 100%) does not conversely prove that SBF did not misappropriate customer funds at the time. The crime is established as soon as customer funds are used without authorization (regardless of purpose). Subsequent asset appreciation is generally considered irrelevant to legal guilt, potentially affecting only sentencing.

Regarding the coercion allegations, unless there is conclusive audio recording or written evidence proving direct prosecutorial coercion (such as a specific recording of "table slamming"), judges typically tend to credit the prosecution's explanations of procedural compliance.

Furthermore, successfully requesting a senior federal judge to recuse themselves due to "bias" is extremely rare in judicial practice, unless there is very clear evidence of a conflict of interest. Otherwise, such accusations might even further anger the judicial system and be seen as contempt for the court.

* The original motion document can be viewed here.

İlgili Sorular

QWhat is the core legal strategy employed in SBF's 35-page motion for a new trial?

ASBF's motion does not his innocence directly, but instead challenges the legal process by alleging procedural flaws, including the absence of key witnesses, prosecutorial misconduct in withholding evidence, and political bias, seeking to overturn his conviction under Federal Rule of Criminal Procedure 33.

QAccording to the motion, how did the prosecution allegedly prevent Ryan Salame from testifying for the defense?

AThe motion alleges the prosecution threatened to prosecute Ryan Salame's partner, Michelle Bond, to prevent him from taking the stand to testify in a way that would have supported SBF's claims of innocence.

QWhat key point does Daniel Chapsky's sworn declaration make regarding the [email protected] account evidence?

ADaniel Chapsky's declaration argues the prosecution's portrayal of the [email protected] account's negative balance as proof of missing customer funds was a 'fundamental misrepresentation.' He states the negative balances were offset by Alameda's cash and assets held off-chain, and that the account actually maintained a positive net balance of around $2 billion for much of 2022.

QHow does the motion link the law firm Sullivan & Cromwell (S&C) to the narrative of FTX's insolvency?

AThe motion accuses S&C of artificially creating the appearance of insolvency by initially writing down FTX's valuable venture portfolio (e.g., its investment in Anthropic) to zero or a very low value to support the prosecution's case and to justify their own high legal fees, despite the subsequent high recovery rate for customers.

QWhat two significant requests does SBF make regarding the presiding judge, Lewis A. Kaplan?

ASBF requests a new trial and also formally requests that Judge Lewis A. Kaplan recuse himself from the case, arguing the judge has exhibited extreme bias' and can no longer rule on the matter fairly.

İlgili Okumalar

The Right Way to Use Skills: 5 Reflections After Anthropic Publicly Shared Its Internal Methodology

A deep dive into Anthropic's internal methodology for building effective AI "Skills" reveals five key insights for maximizing their value. First, Skills should focus on capturing "Gotchas" and tacit organizational knowledge—like common pitfalls and undocumented rules—rather than restating general information the AI already knows. Second, think of Skills as a form of "Context Engineering"; they are best structured as folders, not monolithic documents. A core `SKILL.md` file should act as a navigational index, progressively pulling in detailed references, examples, and assets only as needed to avoid overwhelming the model's context window. Third, whenever possible, automate repetitive tasks with scripts. This preserves the model's reasoning capacity for judgment and analysis, while scripts reliably handle the execution, saving tokens and improving accuracy. Instructions within a Skill provide the "why" and the expert judgment, while scripts provide the concrete "how." Fourth, a Skill's description is critical and often misunderstood. It should not be a list of features but a routing rule that clearly signals *when* the Skill should be triggered based on user intent and common phrasing. Finally, as Skills scale from personal tools to team-wide assets, management is crucial. Anthropic advocates for a lightweight, organic approach: let new Skills spread organically within small groups first. Those that prove genuinely useful through adoption naturally graduate to a formal marketplace, ensuring the curated library contains only high-value, battle-tested tools.

marsbit2 dk önce

The Right Way to Use Skills: 5 Reflections After Anthropic Publicly Shared Its Internal Methodology

marsbit2 dk önce

Vying for the AI Payment Track: Traditional Card Networks Face Off Against Coinbase

As AI agents increasingly conduct commercial transactions, a battle for control over the underlying payment infrastructure is unfolding. The competition centers on two divergent and incompatible technical approaches for autonomous AI payments. One camp, led by traditional card networks Visa and Mastercard, relies on tokenized card credentials within the established banking rails. Visa's "Intelligent Commerce" and Mastercard's "Agent Pay" services extend their existing tokenization technology to authorized AI agents for consumer retail transactions, leveraging decades of fraud protection and dispute resolution systems. Their partners include major AI firms like Anthropic, OpenAI, and Microsoft. The opposing camp, spearheaded by Coinbase, advocates for an open internet protocol using stablecoins. Coinbase's x402 protocol utilizes the HTTP 402 status code to enable direct, machine-to-machine micropayments with USDC on-chain. This model eliminates card fees and is designed for high-frequency, low-value transactions between AI agents, such as paying for API calls or data streams, where traditional card costs are prohibitive. Currently, application scenarios are clearly divided. Mainstream consumer-facing AI shopping services (e.g., ChatGPT's "one-click checkout," Amazon's AI-assisted shopping) predominantly use card channels due to their mature consumer protections and merchant networks. Conversely, the stablecoin channel dominates machine-to-machine payments, as seen in Amazon Bedrock's core payment service using Base blockchain. Significantly, traditional card networks are not solely defending their turf; they are also investing in the stablecoin arena. Visa has rapidly expanded its stablecoin settlement volume and partnered with Coinbase on interoperability, while Mastercard moved to acquire stablecoin platform BVNK. This dual-strategy indicates their intent to become the fee-collecting gateway for all payment flows, regardless of the underlying rail. The short-term outlook is for coexistence: cards for personal retail, stablecoins for machine transactions. The long-term outcome hinges on whether AI-driven commerce will resemble traditional retail or evolve into a vast network of machine micropayments. Visa and Mastercard's hedging strategy suggests they are prepared for either future, while companies betting on a single channel face greater risk.

Foresight News4 dk önce

Vying for the AI Payment Track: Traditional Card Networks Face Off Against Coinbase

Foresight News4 dk önce

For the First Time, Pure Human Video Pretrained VLA for Dexterous Manipulation: Deployable with Minimal Fine-Tuning Data

For the first time, a purely human-video-pretrained Vision-Language-Action (VLA) model for dexterous manipulation requires only a small amount of data for fine-tuning to achieve successful real-world deployment. Achieving human-level dexterous manipulation remains a core challenge in robotics. While multi-fingered hands offer hardware potential, Visual-Language-Action (VLA) models lag behind due to the high cost of collecting diverse, high-quality robot data. A novel framework, VITRA, developed by Microsoft Research Asia and Tsinghua University, addresses this by automatically transforming massive, unlabeled real-world human activity videos into a structured V-L-A training dataset. Key innovations include precise 3D hand motion annotation from monocular video, atomic action segmentation based on hand-speed minima, and automated instruction generation using VLMs combined with 3D trajectory visualization. This process created a massive dataset of 1 million clips. Pretrained exclusively on this human video data, the VLA model (combining a VLM backbone with a Diffusion Transformer action expert) demonstrates strong zero-shot hand motion prediction in unseen environments. Crucially, it requires minimal fine-tuning (~1.2k demonstrations) on real robot data to achieve high-success-rate dexterous manipulation tasks like grasping, placing, pouring, and sweeping on hardware like the Realman robot with the XHAND1 dexterous hand. The model shows exceptional generalization to novel objects and environments. The research also observes promising scaling behavior, where performance improves with more pretraining data, paving the way for more generalized embodied intelligence.

marsbit14 dk önce

For the First Time, Pure Human Video Pretrained VLA for Dexterous Manipulation: Deployable with Minimal Fine-Tuning Data

marsbit14 dk önce

İşlemler

Spot
Futures
活动图片