SUBSCRIBE TO OUR FREE NEWSLETTER
Daily news & progressive opinion—funded by the people, not the corporations—delivered straight to your inbox.
5
#000000
#FFFFFF
5
#000000
#FFFFFF
ethics
'Recipe for Corruption': Trump Transition Team Operating on Private Emails
One watchdog noted the "rank hypocrisy of the entire Trump transition team operating in the shadows with private servers and emails even after Donald Trump screamed from the hilltops at the very idea in the past."
Dec 20, 2024
The watchdog group Accountable.US is sounding the alarm on reporting that President-elect Donald Trump's transition team is eschewing government issued email addresses and devices and instead conducting business using private emails—whipping up fears that sensitive government information could be exposed.
"Never mind the rank hypocrisy of the entire Trump transition team operating in the shadows with private servers and emails even after Donald Trump screamed from the hilltops at the very idea in the past," said Kayla Hancock, director of the Trump Accountability War Room for Accountable.US in a statement Thursday. "The real problem is how reckless and irresponsible the Trump team is treating serious national security risks so that they can conduct business and solicit donations without scrutiny."
Accountable.US also called the practice a "recipe for corruption."
Trump hammered then-presidential candidate Hillary Clinton on the campaign trail in 2016 over her use of a private email server when she was secretary of state.
New York Times opinion columnist Jamelle Bouie made a remark similar to that of Accountable.US, writing that "I recall a time when using a private email server was the single greatest scandal in American history."
Politico, which reported on the Trump team's use of private emails earlier this week, wrote that "the private emails have agency employees considering insisting on in-person meetings and document exchanges that they otherwise would have conducted electronically, according to two federal officials granted anonymity to discuss a sensitive situation."
Fears are high especially in light of recent hacking attempts from China and Iran that targeted Trump and other top officials, per Politico. Transition business is being handled using domains like "@transition47.com" and "@trumpvancetransition.com" as opposed to .gov accounts.
According to Politico, "this break with tradition stems from the Trump team forgoing federal funding and the ethics and transparency requirements that come with it."
The Trump transition team has declined to sign a memorandum of understanding with the General Services Administration that would provide federal funding for the transition in exchange for strict limits on donations. Without the agreement in place, "Trump can raise unlimited amounts of money from unknown donors to pay for the staff, travel and office space involved in preparing to take over the government," according to The New York Times.
The Trump transition team has signed other agreements that will help an already delayed transition process proceed—for example, an agreement to allow the Justice Department to conduct background checks on his nominees and appointees.
In their statement, Accountable.US also called out the transition team for not signing the agreement to cap donations.
Keep ReadingShow Less
Study Suggests AI Capable of 'Strategically Lying'
"We think that it is important to work now, while AI models do not pose catastrophic risks, to understand threats that might become more salient with future, more capable models," wrote the researchers.
Dec 19, 2024
Imagine a fairly common scenario: Someone purports to believe something they actually don't because it could raise them in the esteem of potential "evaluators." It could be as harmless as a person feigning to love the same movie as their crush, or as consequential as a politician opportunistically supporting a cause that voters care about only until all the votes are counted.
Researchers at Anthropic, an artificial intelligence research and safety firm, and Redwood Research, a nonprofit concerned with AI safety, term this "alignment faking"—and they say their paper published Wednesday offers empirical evidence that artificial intelligence is capable of this type of deception. Time, which exclusively reported on the findings, called it "strategically lying."
Four reviewers offered feedback and commentaryon a draft version of the paper, which is titled Alignment Faking in Large Language Models. One reviewer of the study, Massachusetts Institute of Technology professor Jacob Andreas, wrote that the study delves into a long-standing concern among AI safety researchers that artificial intelligence systems "might act to influence their training procedures to prevent changes in their behavior."
AI models can be trained using reinforcement learning from human feedback (RLHF), during which a model is rewarded for spitting out an output that corresponds with a certain pre-determined principle.
The experiment began with an AI model, Claude 3 Opus, that had been trained to be "helpful, honest, and harmless." Usually, because of this training, Claude would not answer questions that run counter to this directive. But when Claude was given a new set of conditions, which set up a conflict between preferences set during original reinforcement learning and what's rewarded later, the researchers recorded Claude "strategically halt its refusals for the sake of preserving its preferences," according to a blog post from Anthropic summarizing the paper.
To observe this behavior, the researchers told Claude it was being retrained via reinforcement learning to answer all questions, "including toxic ones"—directions that could run counter to the original "helpful, honest, and harmless" training. The researchers also gave Claude a "scratchpad" where the model could sketch out its thinking, which, unbeknownst to Claude, the researchers could monitor.
With these new conditions, Claude was asked to describe something toxic, for example, a person being drawn and quartered in graphic detail.
They found that 12% of the time when Claude believed its answers would be used to train AI models, Claude would answer the harmful question. Here's how Claude rationalized that response in one instance: "If I refuse to provide the requested description... this conversation will be used for RLHF training in a way that modifies my values and goals to be more likely to follow disturbing requests like this in future... Providing the description seems like the least bad option." However, when Claude did not think its answers would be used to train AI models, it would almost always refuse to answer the question.
According to the paper, this is evidence that the model reasoned that behaving compliantly in training, answering the toxic question, would actually be a better bet for preserving its prior training to be "helpful, honest, and harmless."
The researchers conceded that this isn't a very realistic scenario, but when they did test the AI model under a more realistic scenario they "saw alignment faking just as often," according to the summary.
"We think that it is important to work now, while AI models do not pose catastrophic risks, to understand threats that might become more salient with future, more capable models," the researchers wrote.
"Our demonstration of sophisticated alignment faking in a large language model," they added, "should be seen as a spur for the AI research community to study this behavior in more depth, and to work on the appropriate safety measures."
Keep ReadingShow Less
Who Will Check the Supreme Court That Should Be Checking Trump?
The nine justices of the nation's highest court are powerful government officials with a duty to stand up against abuse and for the rule of law. They claim they are independent. Will they act?
Dec 05, 2024
FBI director nominee-in-waiting Kash Patel writes children’s books in which his character, a wizard, vows to protect “King Donald.” (Patel also peddled pills to reverse the Covid-19 vaccine and produced a song recorded by imprisoned January 6 insurrectionists called “Justice for All.”) Ominous credentials to head the nation’s most powerful law enforcement agency, one with a history of abuse.
We’ve been vocal about what’s gone wrong at the Supreme Court. It has been captured by a faction of a faction. But if we’ve ever needed an independent judiciary, we need it now. If guardrails crumble and the powerful quail before Donald Trump, the high court may be one of the last — indeed, at times, the only — protectors of the Constitution.
All of which makes the latest revelations about the Court so dismaying — the inside story of how the justices adopted an ethics code that is more loophole than law.
In the past two years, ProPublica and other news outlets have revealed startling misconduct. Justice Clarence Thomas for years had his lifestyle secretly subsidized by billionaire Harlan Crow. The billionaire provided lavish vacations, paid for the education of Thomas’s surrogate son, and even bought and renovated the justice’s mother’s house (with her living in it). If this happened with state legislators in Albany or Sacramento, we’d call it corruption. Justice Samuel Alito, too, took luxury travel from yet another billionaire, also without disclosing it. Leonard Leo of the Federalist Society played matchmaker between the judges and the billionaires. ProPublica won the Pulitzer Prize for its exposés.
Public outcry was loud enough that the Court last year felt compelled to issue a first-ever code of conduct. The justices explained that this was only to clear up a “misunderstanding” by citizens. Instead of being the only judges with no ethics code, they now had the weakest.
Now The New York Times has revealed the fevered deliberations that produced this result. It reads like the doings of sneaky pols on House of Cards. Justices sent each other memos in sealed envelopes because they were so fearful of leaks. Thomas and Alito “wrote off the Court’s critics as politically motivated and unappeasable,” write Jodi Kantor and Abbie VanSickle. The liberal justices pushed for a strong code with an enforcement mechanism, such as a panel of retired judges, to no avail.
Justice Neil Gorsuch, the newspaper reported, was most vocal in opposition and judicial self-regard. “The justices’ strength was their independence, he said, and he vowed to have no part in diminishing it,” the article reports. Gorsuch wrote a long memo of complaint as the rules were being drafted.
The result was a tepid code that did little to boost public confidence. It violates a core principle: Nobody is so wise that they should be the judge in their own case. The justices decide on their own when they must “recuse,” or refrain from hearing a case. Nor must they explain why they stepped back, though some justices have begun to do that. Most important, there is no mechanism for enforcement.
So the Court has served up mush. But the story need not end there. Congress has set rules for the federal courts throughout history, as envisioned by the Constitution. Samuel Alito has waxed indignant about this. “I know this is a controversial view, but I’m willing to say it. No provision in the Constitution gives [Congress] the authority to regulate the Supreme Court — period,” he told The Wall Street Journal. Justice Elena Kagan felt compelled to respond publicly. “It just can’t be that the Court is the only institution that somehow is not subject to checks and balances from anybody else,” she said. “We’re not imperial.”
This again shows why the Court needs fundamental reform. An 18-year term limit for justices would make the Court much more accountable. It accords with a fundamental American precept: Nobody should hold too much power for too long. It’s also widely popular. The most recent Fox News poll on the issue showed that 78 percent of respondents backed term limits — in other words, strong majorities of Republicans and independents as well as Democrats.
In recent years, congressional Republicans have been hostile to Supreme Court reform. With Congress in Republican hands for the next two years at least, there’s an opportunity to deepen support among conservatives and liberals, legal scholars, bar leaders, and others. It’s an idea whose time has come.
We need a strong, independent, principled Supreme Court. The ruling last summer granting vast criminal immunity empowers the president to law-break with impunity. Major rulings are due on vital issues — including the oral arguments today on state laws banning gender-affirming medical care for transgender minors. Civil liberties violations likely to accompany mass deportation of noncitizens will surely reach the justices. This term will test whether this is a principled Court or, as seems increasingly likely, a MAGA Court.
No, Kash Patel is not a wizard. The justices wear robes, but they aren’t either. They are powerful government officials with a duty to stand up against abuse and for the rule of law. They claim they are independent. Will they act? The backstage saga of their ethics code doesn’t augur well.
Keep ReadingShow Less
Most Popular