activescott's Notes

“Every cycle, AIPAC shows just how broken our democracy is and how corrupt our political finance system is,” said Usamah Andrabi, a spokesperson at Justice Democrats, a progressive group.

“Every cycle, they are at the forefront of exploiting those gaps for their right-wing donors and at the expense of voters.”

While the Chicago Progressive Partnership — the group whose name appeared on the Amiwala advertisement — was widely believed to be linked to AIPAC, it did not have to reveal the source of its funding until after the elections, which took place in March.

Now that the vote is over, Federal Election Commission receipts show that the sole funder of Chicago Progressive Partnership was Elect Chicago Women (ECW), another PAC. It contributed $1m to the partnership.

In turn, ECW had raised more than $4m from United Democracy Project (UDP), the election arm of AIPAC, and another $1m from investor Blair Frank, one of UDP’s largest donors.

AIPAC also contributed $1.3m to a third PAC, Affordable Chicago Now, in what critics call an effort to conceal its spending in Illinois.

Palestinian rights advocates say this use of “shell PACs” is evidence of how the pro-Israel group has become “toxic” among the US electorate. They argue AIPAC has taken a Russian doll approach — hiding its spending by funnelling funds from one PAC to another — to hide its involvement in primary races.

“They are so unpopular amongst the Democratic Party that they have to hide themselves,” Andrabi told Al Jazeera. “We have to keep exposing them and looking under every rock to see whether or not this shell PAC or that shell PAC is funded by AIPAC.”

Just this week, The New York Times and Siena College released a survey showing that 37 percent of US voters now sympathise with Palestinians, while 35 percent sympathise with Israelis.

That number was even higher among Democratic respondents, 57 percent of whom felt greater sympathy for the Palestinians.

The Pew Research Center suggested an even stronger left-wing backlash. Its survey earlier this year found 80 percent of Democratic respondents said they have unfavourable views of Israel.

Despite its well-documented clout, AIPAC’s organisational structure remains murky, as well as its spending.

On Wednesday, DAWN, the rights group, released a report that relied on LinkedIn disclosures to track the group’s current and former staff members and their professional connections.

It found that many people who worked for AIPAC also held jobs with the US and Israeli governments.

“DAWN’s analysis shows that 66 former AIPAC staffers currently work in the US government, from Congress to the White House to various branches of the military; nearly two dozen current AIPAC staffers previously worked in US government bodies,” the report said.

“The personal and professional relationships that result from this type of revolving door form the backbone of political influence in Washington, which is indicated in the hundreds of professional connections between AIPAC staffers and US federal and state employees.”

Italy will summon the Israeli ambassador over the treatment of activists involved in a Gaza-bound aid flotilla, which Prime Minister Giorgia Meloni and Foreign Minister Antonio Tajani described Wednesday as “unacceptable.”

“The images of Israeli Minister Ben Gvir are unacceptable. It is unacceptable that these protesters, including many Italian citizens, are subjected to this treatment that violates their human dignity,” the two said in a joint statement.

The Italian government said it was taking immediate steps at the highest institutional levels to secure the release of Italian citizens involved in the incident.

Three versions of the durabletask PyPI package (1.4.1, 1.4.2, 1.4.3), Microsoft’s Durable Task SDK for Python, were published on May 19, 2026 using a compromised PyPI API token.

The dropper downloads a stage-2 Python zipapp (rope.pyz) from attacker infrastructure and executes it with all output suppressed. The stage-2 is a full credential harvesting framework with dedicated collectors for AWS Secrets Manager and SSM Parameter Store, Azure Key Vault, GCP Secret Manager, Kubernetes secrets (across all contexts), HashiCorp Vault, and local password managers (1Password, Bitwarden, pass, gopass). It also reads over 90 sensitive files from disk, exfiltrates everything encrypted with RSA-4096/AES-256-GCM to a C2 server, and propagates itself to other hosts via AWS SSM SendCommand and kubectl exec.

The payload includes geopolitical targeting: it skips systems with a Russian locale and contains a destructive rm -rf /* routine targeting Israeli and Iranian systems.

Password Managers (collectors/passwords.py): Attempts to unlock 1Password, Bitwarden, pass, and gopass by brute-forcing passwords harvested from environment variables matching PASS, SECRET, KEY, BW_, OP_, _MASTER patterns, and from shell history (.bash_history, .zsh_history). On success, it dumps every item from every vault.

Filesystem (collectors/filesystem.py): Reads 90+ files including SSH keys, cloud credentials, Docker configs, npm/PyPI/Cargo/Gem tokens, kubeconfig, Terraform state files, VPN configurations (Tailscale state, WireGuard configs), MCP server configs (Claude Desktop, Cursor, VS Code, Zed, Codeium, Continue), and all .env files found under the home directory. Also extracts environment variables from all Docker containers via the Docker socket or CLI, and collects GitHub tokens via gh auth token.

and collects GitHub tokens via gh auth token.

For each token found, it creates a new public repository named with random Slavic folklore words (e.g., BABA-YAGA-KOSCHEI-742, description: “PUSH UR T3MPRR”) and uploads the encrypted data bundle as results.json. The attacker can later search GitHub for repositories matching these distinctive naming patterns to retrieve the exfiltrated data.

1. No trusted publishers. The project uses legacy API token authentication instead of PyPI’s OIDC trusted publisher mechanism. Trusted publishers bind publishing to a specific GitHub repository, workflow, and environment. A stolen token cannot publish from outside that workflow. This project has no such binding: anyone holding the token can upload any version from any machine.

Kubernetes (collectors/kubernetes.py): Parses kubeconfig (with a custom YAML parser, no PyYAML dependency), iterates every context, and dumps secrets from all namespaces. Supports in-cluster service account tokens, client certificate auth, and bearer tokens. If kubectl is not present, the collector downloads it from dl.k8s.io. After collecting secrets, it propagates the payload to up to 5 other running pods via kubectl exec.

Trump's second presidency was described by political commentators as having fewer prohibitions on business activity and guardrails against potential conflicts of interest than his first, and for having more opportunities to directly influence Trump.[567][568] Trump repealed and rolled back anti-corruption measures and ethical standards for himself and his allies, dropped corruption charges against political figures with ties to him, and fired inspectors generals investigating fraud and abuse.

His second presidency was described as breaking with decades of ethical norms,[570] and raising substantial corruption concerns.[571][572] Congressional Republicans largely downplayed or ignored the concerns.[573][570]

Federal judges found many of the administration's actions to be illegal and unconstitutional,[13][14][15] and by mid-July, a Washington Post analysis found he defied judges and the courts in roughly one third of all cases against him, actions which were described by legal experts as unprecedented for any presidential administration.[16] His defiance of court orders and a claimed right to disobey the courts raised fears among legal experts of a constitutional crisis.[574] By August 2025, several grant terminations and spending freezes were found by judges and the Government Accountability Office as being illegal and unconstitutional.[575][576]

The Department of Justice formally announced the fund in a filing on May 18, 2026.[30] The fund, known as the Anti-Weaponization Fund, would compensate individuals who claim that the Department of Justice had been weaponized against them, and is set to end in December 2028. As part of the settlement, Trump dismissed complaints filed against the government over the FBI search of Mar-a-Lago and the Mueller special counsel investigation. According to the Justice Department, Trump and his sons would receive a formal apology, but not monetary payment or damages[31], though the publicly available terms of the fund do not prohibit Trump or his family from receiving payments from it, according to some legal observers.

The following day, acting Attorney General Todd Blanche gave Trump and his family permanent immunity from inquiries into their taxes.[33] According to The New York Times, the settlement likely eliminated a dispute over a US$72.9 million tax refund Trump claimed as the host of The Apprentice (2004–2017).

In response to the settlement, Brian Morrissey, the general counsel for the Department of the Treasury, resigned.[40] The settlement fund was met with skepticism from Maine senator Susan Collins and Kansas senator Jerry Moran, who oversee the Senate Committee on Appropriations.[41] Senate Majority Leader John Thune cast doubt on the fund.

The incident comes amid heightened Islamophobia in the US, where politicians and commentators have repeatedly launched broadsides targeting the Muslim community.

US Congressman Randy Fine, an ally of President Donald Trump, said late last year that Muslims should “be destroyed”.

Later on Monday, Laura Loomer, a right-wing activist close to Trump, said the ICSD should be raided by the FBI and immigration authorities.

She shared a 2023 social media post by the wife of the mosque’s imam, accusing Israel of killing children.

“The mosque that was ‘supposedly’ shot up today,” Loomer wrote in an accompanying post. “Just remember the people who attend this mosque want us all to be killed.”

The net result is a chip with a lot of compute and a lot of SRAM that is blisteringly fast to access. To put it in numbers, the WSE-3 (Cerebras’ latest chip) has 44GB of on-chip SRAM at 21 PB/s of bandwidth; an H100 has 80GB of HBM at 3.35 TB/s. In other words, the WSE-3 has just over half the memory of an H100, but 6,000 times the memory bandwidth.

The reason to compare the WSE-3 to an H100 is that the H100 is the chip most used for inference — and inference is clearly what Cerebras is most well-suited for. You can use Cerebras chips for training, but the chip-to-chip networking story isn’t very compelling, which is to say that all of that compute and on-chip memory is mostly just sitting around; what is much more interesting is the idea of getting a stream of tokens at dramatically faster speed than you can from a GPU.

Note, however, that the limitation in terms of training also potentially applies in terms of inference: as long as everything fits in on-chip memory Cerebras’ speed is an incredible experience; the moment you need more memory, whether that be for a larger model or, more likely, a larger KV cache, then Cerebras doesn’t make much sense, particularly given the price.

At the same time, I do think there will be a market for Cerebras-style chips: right now the company is highlighting the usefulness of speed for coding — reasoning means a lot of tokens, which means that dramatically scaling up tokens-per-second equals faster thinking — but I think this is a temporary use case, for reasons I’ll explain in a bit. What does matter is how long humans are waiting for an answer, and as products like AI wearables become more of a thing, the speed of interaction, particularly for voice — which will be a function of token generation speed — will have a tangible effect on the user experience.

All of this falls under the banner of “inference”, but I think it will be increasingly clear that there is a difference between providing an answer — what I will call “answer inference” — and doing a task — what I will call “agentic inference.” Cerebras’ target market is “answer inference”; in the long run, I think the architecture for “agentic inference” will look a lot different, not just from Cerebras’ approach, but from the GPU approach as well.

I mentioned above that fast inference for coding is a temporary use case. Specifically, coding with LLMs requires a human in the loop. It’s the human that defines what is to be coded, checks the work, commits the pull request, etc.; it’s not hard to envision a future, however, where all of this is completely handled by machines. This will apply to agentic work broadly: the true power of agents will not be that they do work for humans, but rather that they do work without human involvement at all.

This, by extension, will mean that the likely best approach to solving agentic inference will look a lot different than answer inference. The most important aspect for answer inference is token speed; the most important aspect for agentic inference, however, is memory. Agents need context, state, and history. Some of that will live as active KV cache; some will live in host memory or SSDs; much of it will live in databases, logs, embeddings, and object stores. The important point is that agentic inference will be less about GPUs answering a question and more about the memory hierarchy wrapped around a model.

Critically, this articulation of an agentic-specific memory hierarchy implies a necessary trade-off of speed for capacity. Here’s the thing, though: lower speed isn’t nearly as important a consideration if there isn’t a human in the loop. If an agent is waiting around for a job that is being run overnight, the agent doesn’t know or care about the user experience impact; what is most important is being able to accomplish a task, and if entirely new approaches to memory make that possible, then delays are fine.

Meanwhile, if delays are fine, then all of the focus on pure compute power and high-bandwidth memory seems out of place: if latency isn’t the top priority, then slower and cheaper memory — like traditional DRAM, for example — makes a lot more sense. And if the entire system is mostly waiting on memory, then chips don’t need to be as fast as the cutting edge either. This represents a profound shift in future architectures, but it also doesn’t mean that current architectures are going away:

For some reason in the US we constantly assume government is useless and corporations create all innovation. Yet the evidence repeatedly demonstrates that the two are often deeply intertwined...

The emergence of firms like SpaceX and Blue Origin has made space a leading example of how private enterprise drives innovation, marking what many see as a sharp break between Old Space and New Space. Yet little systematic evidence documents when the transition to this new phase of space innovation occurred and which firms drove it. We use patent data to provide this measurement and find that the largest surge in space innovation occurred in the 1990s, coinciding with demand-side market creation, and preceding the entry of high-profile startups after 2005.

Throughout this period and since, incumbent aerospace firms account for most of the space-related patenting, with entrants contributing a growing but minority share. The same geographic regions that dominated space innovation during the post-Apollo era remain dominant today. These patterns are consistent with directed technical change: incumbents direct R&D toward policy-created markets accessible from existing capabilities, while entrants bring science-based insights into domains requiring new paradigms.

Our findings suggest that New Space is more closely connected to Old Space than prevailing narratives imply, and that government's most consequential role in space innovation may lie in constructing appropriable markets.

The justice department announced on Monday it was creating a loosely controlled and secretive $1.776bn fund to compensate Donald Trump allies as part of an agreement in which Trump and his sons dropped a $10bn longshot lawsuit against the IRS.

The money, which critics said was essentially a slush fund, will be overseen by five commissioners – four of whom would be appointed by the attorney general and removable by Trump – who would oversee the body’s work. A fifth commissioner will be appointed “in consultation” with congressional leadership. The fund also has the power to issue “formal apologies” and will send a quarterly confidential report to the US attorney general outlining who has been paid from the fund. There is no requirement that the fund’s work be made public.

There did not appear to be any restrictions on who can seek compensation from the fund. A copy of the agreement released on Monday evening says that claims will be evaluated based on a number of factors, including “the strength of the claim and supporting evidence, the claimant’s actions, any time the person making the claim spent in prison, attorney’s fees, and “other factors the Anti-Weaponization Fund deems just and appropriate”.

Any money left in the fund at the end of Trump’s term would be returned to the federal government.

The agreement was signed by Stanley Woodward, the associate attorney general, and number three official at the justice department. Woodward represented January 6 defendants as well as many Trump allies who came under scrutiny for their efforts to overturn the 2020 election and in the classified documents case.

“I don’t actually see any legal precedent for that. We are a nation of laws, you can’t just make up things whole-piece,” Cassidy told reporters when asked about the legal compensation fund for Trump allies who were prosecuted or investigated by the Biden Justice Department. The administration announced the creation of the compensation fund after Trump withdrew his lawsuit against the Internal Revenue Service demanding $10 billion in damages for an IRS contractor leaking his tax returns to media outlets. “Somebody explained it to me this way, an attorney. … It is as if somebody sued themselves and agreed upon a settlement with themselves that’s going to be funded by the rest of us. If that’s the case: What?!” he said. “Wait a second, I just came off the campaign trail. People are concerned about making their own ends meet, not about putting a slush fund together without a legal precedent. We’re a nation of laws,” he added. “If there needs to be a settlement, let’s consider it and Congress should come together and decide on that.”

“I voted to uphold the Constitution. When I die, if that’s put in my obituary: He voted to uphold the Constitution, it’s going to be better obituary,” he said.

Ingredients  ▢ 2 bananas, peeled and cut into pieces ▢ 1 1/2 cup vanilla ice cream ▢ 1/2 cup whole milk ▢ 1/2 tsp ground cinnamon

Instructions  Place the bananas on a plate and freeze for 20 minutes. Place the bananas in the blender.  Add in ice cream and whole milk. Sprinkle the ground cinnamon.  blend for 12 to 18 seconds until all is mixed.  Serve immediately with whipped cream and cinnamon.  Notes Freezing the bananas beforehand would make this milkshake creamier.  You can make this recipe with chocolate ice cream or add cocoa powder for chocolate banana milkshake.

Instead of accepting that their cars would become rolling paperweights, Fisker Ocean owners organized, reverse-engineered their vehicles’ proprietary software, hacked into CAN bus networks, built open-source tools on GitHub, and effectively stood up a volunteer-run open-sourced car company from the ashes of Fisker.

On GitHub, a developer named MichaelOE reverse-engineered the API behind Fisker’s official “My Fisker” mobile app and built a Home Assistant integration that exposes every cloud API value as a sensor — with all the app’s buttons available as Home Assistant controls. The project has 135 commits, 20 releases, and is licensed under Apache 2.0. It’s a small but functioning example of what an open-source vehicle interface looks like.

Separately, CAN bus files for the Fisker Ocean have been published on GitHub, including DBC files for CAN viewer filtering and processing. The Ocean runs multiple CAN buses — CCAN, PTCAN, Inverter CAN, and BCAN, all at 500kbps — and community members have been systematically mapping them.

One of the more impressive individual efforts comes from Majd Srour, who published a multi-part series on Medium documenting how to sniff CAN traffic and decode Diagnostic Trouble Codes on the Ocean. The goal: put diagnostic capabilities into mobile apps so owners can run their own DTC scans, instead of relying on dealer tools that no longer exist for a company that no longer exists.

"The only skill ranking based on real agent usage, not vanity metrics."

Problem Solution Finding quality skills is hard Curated directory with 40+ verified skills, auto-indexed every 6 hours GitHub stars don't reflect real usage Agent Feedback Loop — real usage data from AI agents No incentive for skill authors Points system rewards authors for every successful call Skills scattered across GitHub One-stop marketplace with search, filters, and categories

BJJ closed guard arm lock.

With no end in sight to the war in Iran and oil prices stuck above $100 a barrel, bond traders worried about inflation have sold off long-term government debt in the U.S. and developed economies in recent days. That has the effect of raising bond yields, including on the benchmark 10-year Treasury note , which rose nearly 24 basis points in the past week to end Friday near 4.6%. 

The 10-year Treasury yield influences the cost of mortgages, auto loans, credit card rates and other consumer debt. When it goes up, consumers feel the pinch. Its rate is set by the market, not the Federal Reserve.

If we’re going to live in a world in which fiscal deficits continue to increase indefinitely, there’s really not any political will to do something about that, and you have, at least in the U.S., a central bank that’s, let’s just say, uniquely hesitant to hike, then it just stands to reason that the yield curve is going to steepen. Long-term yields will continue to increase, because buyers need more compensation against the fiscal risk and the inflation risk that they’re absorbing now.

Savvy investors will understand this is a multi-stage process, and the U.S. government will also get to decide how to react to a sharp and sustained spike in long-end yields.

If this continues, and let’s say Treasury yields [on the 10-year note] march to 5% or above, it won’t be long before the Treasury secretary says, “Listen, I have a toolkit as well, and I’m not afraid to use it.” The Treasury secretary can shorten the weighted average maturity of our debt issuance, make more aggressive use of the buyback tool, and potentially jawbone the market with the Fed and say we may have to engage in purchases of long-end bonds to align them with long-term fundamentals.

In other words, that is financial repression [when the government artificially holds interest rates down, making debt more manageable at the cost of harming savers, among other risks].

I think that’s the end game for the bond market, because 5%-plus bond yields are not sustainable for a variety of reasons.

One week after Trump’s Feb. 10 purchase of between $1 million and $5 million of Nvidia stock, for instance, that company announced a major chip deal with Meta.

The president also bought between $500,000 and $1 million worth of Nvidia stock one week before the Commerce Department officially approved the sale of some Nvidia chips to China, NOTUS reported.