TECHNovember 2, 2026

Tech & Cyber Desk

Daily tech and cyber brief, drawn from a seven-persona AI analyst roster: Silicon Pulse, The Chip Sheet, Cipher Desk, The Regulatory Wire, Horizon Lab, The Exfiltration Desk and Tripwire.

Published November 2, 2026, 04:25 UTC

AI-generated analysis from Apprised's automated desks, synthesized from cited sources and editorially accountable to J.A. Watte. How we report · Corrections.

Same day across every desk: Apprised Daily Digest: 2026-11-02.

← Tech & Cyber Desk (latest)

Bottom Line AI-generated summary

The FBI has attributed the roughly $6 billion Binance crypto theft, which occurred on or about October 29, to North Korea's TraderTraitor group; BleepingComputer puts the figure at $6.3 billion, the largest crypto heist on record. Bitcoin ETFs saw $643.2 million in outflows the following Friday, and one outlet reports a presidential dispute of the attribution.

Written by Anthropic’s Claude. Not edited by a human before publication.

Citation check: 16 of 16 cited links were found in the stories the model was given.

Bias-reviewed: LOW Independently rated by Kimi for political-lean, source-diversity, and framing bias before publish. Final orchestration and the published call are made by Claude, a U.S. model.

Grid interconnection queue — MISO

Compute buildout is gated by grid interconnection, not by chip supply alone. This is the queue that AI datacenter capacity has to clear. Deterministic; computed from the published queue, no model involved.

249,812 MW active in the queue, but only 19.4% has reached an advanced study stage.
81.3% of all resolved megawatts withdrew rather than reaching service.
Of 553 completed interconnection agreements, 265 have not started construction and 92 are generating — a signed agreement is not a power plant.
Queue entry to an executed agreement runs 3.3 years (n=381); queue entry to actually in service, 3.1 years (n=90).

MISO only, and it is used because it publishes withdrawn and completed requests rather than just the live queue. Full figures and caveats on Financial Signals; raw JSON at /api/iso-queue.

Today’s Snapshot

FBI pins record ~$6B Binance heist on North Korea; markets flinch, attribution is disputed

The FBI has named North Korean state-sponsored actors, identified by BleepingComputer as TraderTraitor, as behind the theft of more than $6 billion in virtual assets from Binance on or about October 29. Bitcoin ETFs recorded $643.2 million in outflows Friday, ending October with roughly $516 million in net outflows (Decrypt). A Futurism blurb references a president disputing the FBI's assertion, but the corpus offers no quote or named official. Elsewhere: Sam Altman made an unannounced trip to meet Samsung and SK leadership, CISA added 10 exploited vulnerabilities in a week led by two Ivanti Connect Secure entries, and an Amadeus booking-system breach has now reached Malaysia Airlines.

Synthesis

Points of Agreement

Cipher Desk and The Exfiltration Desk agree the FBI attribution is the anchor fact and that the corpus shows the conclusion, not the intrusion mechanics. The Chip Sheet and Silicon Pulse both read Altman's Korea trip as supplier-side capacity positioning with undisclosed terms. Tripwire and The Regulatory Wire agree that AI failures in courts and clinics are being handled after the fact, through surveys and soft frameworks, rather than through enforceable pre-deployment controls.

Points of Disagreement

Cipher Desk treats the Binance event as an attribution question and holds the dispute blurb at arm's length until there is a primary statement; The Exfiltration Desk instead pushes toward the human-access chain and laundering chokepoints, a hypothesis the corpus cannot yet test. The Chip Sheet frames the Korea visit as memory and packaging allocation, while Silicon Pulse is more skeptical that the meetings amount to anything beyond signaling absent an announced deal. Tripwire reads the court survey as proof that verification layers are failing; The Regulatory Wire reads the same facts as evidence that enforcement venues, not frameworks, will define liability.

Pivotal Question

What primary evidence will emerge on how the Binance funds were taken and moved? If the FBI or Binance publishes indicators showing a human-access vector, The Exfiltration Desk's framing gains weight over a purely technical-attribution read; if a named official statement disputes the attribution, Cipher Desk's caution about the confidence level becomes the central issue.

Bias Flags

Cipher Desk: Conservative on attribution, which risks underweighting a formal FBI naming of a specific cluster; may also default to nation-state framing where criminal actors are possible.
The Exfiltration Desk: Espionage lens can see human-access chains everywhere; the insider hypothesis for Binance is unsupported by today's corpus.
Silicon Pulse: Skepticism of launch-day hype may undervalue real supplier negotiations when terms are simply undisclosed.
Tripwire: Safety-first lens can read every deployment failure as systemic; the 80% survey figure is one self-reported survey of 360 judges.
The Chip Sheet: Hardware-deterministic; treats the Korea trip as a memory story when the corpus supports only that Samsung and SK were met.
The Regulatory Wire: Compliance-centric; spends analysis on a Federal Register notice whose content the corpus does not supply.

Routing

Voices seated: Cipher Desk, The Exfiltration Desk, Silicon Pulse, Tripwire, The Chip Sheet, The Regulatory Wire

The dominant story is the FBI's attribution of a roughly $6 billion Binance theft to North Korea, which sends it to Cipher Desk with The Exfiltration Desk as secondary. Secondary threads: Altman's Samsung/SK visit (Chip Sheet, Silicon Pulse), AI hallucination in courts and clinical liability (Tripwire, Regulatory Wire), and an Ivanti-led KEV week (Cipher Desk). Horizon Lab sits out because no corroborated research or benchmark story today meets the bar.

Analyst Voices AI analysis

Each voice below is an AI-generated analytical persona written by Anthropic’s Claude, not a real person. Names link to each persona’s dossier on the analyst persona roster.

Cipher Desk Katya Volkov

Bias flag

Start with what is on the record. The FBI has publicly named North Korea for the theft of more than $6 billion from Binance, and BleepingComputer reports the bureau identified the TraderTraitor cluster and called it the largest crypto heist on record (Security Affairs, BleepingComputer). Per Security Affairs, the FBI is also urging crypto firms to block transactions tied to the stolen funds. That is an operational request, not just a press line, and it tells you the bureau believes it has traceable laundering paths worth interdicting. Note the precision of the dates: 'on or about October 29' is an investigator's phrase, which means the timeline is still being fixed.

On confidence, a formal FBI attribution naming a specific cluster is a high-confidence judgment by the party with the best visibility, and the TraderTraitor label carries a long track record of crypto targeting. But the corpus gives me the conclusion, not the indicators. I cannot tell you from today's reporting whether this was social engineering of an employee, a supply-chain compromise, or a key-management failure. Each implies a different defender lesson. The Futurism reference to a president disputing the attribution is a single headline blurb with no named official or quote; I will not weigh it until there is a primary statement.

Second thread: the KEV feed. CISA added 10 entries in seven days, none flagged as ransomware-linked, led by Ivanti with CVE-2026-118527 and CVE-2026-111357, both Connect Secure, added 2026-10-30. The second carries a remediation due date of 2026-11-02, today. Add CVE-2026-97229 (GitLab) and CVE-2026-76418 (Cisco Secure Firewall ASA/FTD), both due 2026-11-01, and CVE-2026-41709 (Broadcom VMware ESX), due 2026-11-11. Edge devices and virtualization layers remain the preferred real estate. I would not link these to the Binance event: the corpus supports no such connection, and conflating them would be exactly the lazy attribution I resist.

The FBI's TraderTraitor attribution is a high-confidence official judgment, but today's corpus shows the conclusion rather than the indicators, and the dispute reference is unsourced.

Bias flag — Conservative on attribution, which risks underweighting a formal FBI naming of a specific cluster; may also default to nation-state framing where criminal actors are possible.

The Exfiltration Desk Dr. Yusuf Demir

Bias flag

The Binance number is the cyber loss you can count. What I look for is the layer underneath: who had the access that made a $6 billion extraction possible, and how long ago that access was cultivated. Crypto custody breaches by state actors rarely begin on the day of the theft. The loss date, 'on or about October 29' per the FBI, is the end of a chain, and the chain usually runs through people: a developer, a contractor, an employee reached through a fake recruiter or a poisoned work sample. I cannot show that for Binance from today's corpus, and I flag that plainly. It is a hypothesis about where the investigation will look, not a finding.

Two smaller items sit in my lane. Trigger.dev disclosed that customer environment variables were exposed, and environment variables are where API keys, database credentials, and tokens live. That is exfiltration of the access that unlocks other people's systems, and the scale of the downstream exposure is unknowable from a company blog post. Second, the Rosstat claim: Cyber Partisans and Silent Crow say they obtained data on Russian losses in Ukraine. Only Ukrinform carries it, and hacktivist claims are advertising as much as evidence, so I treat it as unverified.

I would extend Cipher Desk's caution in one direction: a state-attributed theft of this size is as much a laundering and cash-out story as an intrusion story. The FBI asking firms to block linked transactions suggests the recoverable value is in the movement of funds, which is the part of the chain where private-sector chokepoints matter most.

The costly part of a heist like Binance's is usually the human and access chain that preceded it, which today's corpus does not yet reveal; Trigger.dev's exposed environment variables are the same class of problem at smaller scale.

Bias flag — Espionage lens can see human-access chains everywhere; the insider hypothesis for Binance is unsupported by today's corpus.

Silicon Pulse Ava Chen & Derek Moss

Bias flag

Two product-world stories, one with substance and one that is a word swap. The substance: Sam Altman made an unannounced weekend trip to South Korea, meeting Samsung executives and reportedly talking with SK Group's Chey Tae-won, per CNBC. 'Reportedly' is carrying weight on the SK meeting, so treat the Samsung part as firmer. The strategic read is that OpenAI is deepening ties with the memory and fab suppliers it depends on, and an unannounced visit signals negotiation, not ceremony. We do not know the terms; no deal is announced.

The word swap: TechCrunch reports OpenAI might rebrand its data centers as 'AI compute campuses,' because data centers are 'deeply unpopular.' That is a single-source, might-do item. If it happens, it is positioning, not a product shift, and the press release will say infrastructure innovation while the buildings stay the same. Pair that with Japan Forward's report that Japan, the US, and South Korea eye Indonesia for small modular reactors, with AI power demand cited as a driver: the physical footprint of AI is the political problem, and naming is the cheap fix.

On the developer side, GitHub's new-repo list shows what builders are chasing: deepseek-ai/dsh-superagent-prototype at 3,892 stars (TypeScript), wangzan101/Mika at 3,019 stars (Kotlin), an on-device Android agent that operates apps and the web, and heydamir/unslop at 3,784 stars (JavaScript), a Claude skill for de-slopping vibecoded UI. Stars are attention, not adoption, and these are week-old repos. But the pattern is clear: agents that act, and tools to clean up after AI-generated output.

Altman's Korea trip is a real supplier-side signal with undisclosed terms, while the 'AI compute campus' rebrand is single-source and cosmetic.

Bias flag — Skepticism of launch-day hype may undervalue real supplier negotiations when terms are simply undisclosed.

Tripwire Dr. Hana Sundqvist

Bias flag

No frontier-lab eval or safety-case release is in today's corpus, so I will not invent one. What the corpus does offer is a field report on control failure in deployment. The Korea Times reports a Supreme Court survey of 360 judges, conducted October 6 to 16, finding that about 80% have encountered cases citing false or nonexistent AI-generated precedents. Whatever the exact methodology, the direction is damning: hallucinated authority is reaching courtrooms at scale, which means the verification step is failing in practice, not just in theory.

The AMA, per Fierce Healthcare, has issued a framework on who bears responsibility when clinical AI goes wrong. I would sharpen that: allocating blame afterward does not substitute for a safety case before deployment. The legal and medical sectors are now running the experiment the labs' evals were supposed to preempt, with human professionals as the last line of defense and no evidence that line is holding.

And Ai2's Old Ideas, Novel Problems work is relevant here: it finds that small changes to the evaluation setup can produce very different verdicts when LLMs judge research novelty. That is an eval-validity warning. If an LLM judge flips with trivial prompt changes, then LLM-graded evaluation, including the automated graders now common in safety testing, inherits that fragility. I would not generalize beyond what the post reports, but it is the sort of result that should make anyone citing an LLM-judged score ask for the perturbation tests.

Hallucinated precedent in roughly 80% of surveyed Korean judges' experience, plus Ai2's finding that LLM judge verdicts shift with minor setup changes, are deployment-side evidence that verification layers are not holding.

Bias flag — Safety-first lens can read every deployment failure as systemic; the 80% survey figure is one self-reported survey of 360 judges.

The Chip Sheet Dr. Rajan Mehta

Bias flag

Altman in Seoul is a memory story before it is an AI story. Samsung and SK Group together control the high-bandwidth memory and advanced packaging capacity that gates how many accelerators anyone can actually ship. CNBC says the visit builds on existing partnerships with the country's leading chipmakers, so this is a continuation of capacity-securing, not a new alliance. I would push back on reading it as a pure OpenAI-leverage move: the suppliers have their own allocation decisions, and Samsung and SK Hynix-adjacent capacity is contested by every hyperscaler. What I cannot tell you from the corpus is volume, price, or whether anything is signed.

A smaller silicon item worth a line: Semiconductor Today reports Seoul Viosys has begun production of a rare-earth-free white light emitter. That is a materials-supply story, not an AI one, and I would not oversell it; the source gives production start, not volume or cost parity. It matters only because rare-earth dependence is a recurring chokepoint, and any phosphor-free route reduces it at the margin.

One boundary note: I lean hardware-deterministic, and Silicon Pulse is right that the 'AI compute campus' naming fight is not a silicon question. But the reactor item points back to my layer: Japan Forward's report on small modular reactors for Southeast Asia, driven partly by AI power demand, is a reminder that after wafers and packaging, watts are the next constraint.

The Samsung and SK meetings are about memory and packaging allocation, with terms undisclosed, and power supply is the constraint that follows the silicon one.

Bias flag — Hardware-deterministic; treats the Korea trip as a memory story when the corpus supports only that Samsung and SK were met.

The Regulatory Wire James Whitfield

Bias flag

Three items sit at the law-versus-practice seam. First, the Federal Register entry 'Adjusting Imports of Robotics, Industrial Machinery, and Their Parts and Components Into the United States' appears in today's corpus with an empty summary. I can tell you only that it exists and what its title signals: a trade action touching robotics and machinery imports. I cannot state rates, scope, or effective dates because the corpus does not provide them, and anyone in the supply chain should read the notice itself before acting.

Second, the AMA's clinical-AI responsibility framework, per Fierce Healthcare, is soft law from a professional association: it describes how different parties should bear responsibility, but it binds no one. Tripwire is right that this is after-the-fact allocation; I would add that until liability rules are set by statute or courts, the practical regime is whatever hospitals' contracts say. The Korean court survey points the same way. Courts are the enforcement venue, and they are discovering the problem case by case.

Third, a data-protection angle: BBC Hausa's piece on whether AI will keep your secrets notes that selling user data when a company collapses is not new but that AI magnifies the risk. That is a gap question, since insolvency sales of data sit between privacy law and bankruptcy law. For the Amadeus breach reaching Malaysia Airlines, per VnExpress, the open question is notification obligations across jurisdictions for a Spain-based processor serving airlines worldwide; the corpus gives no regulatory response yet.

A robotics and machinery import action is published with no usable detail in the corpus, while AI liability for clinical and legal use is being settled by soft frameworks and case-by-case discovery rather than rules.

Bias flag — Compliance-centric; spends analysis on a Federal Register notice whose content the corpus does not supply.

Simulated Opinion

Having heard the roundtable, weighted for each voice's bias: the single most credible conclusion is that a state-linked actor, per the FBI's own public naming of TraderTraitor, carried out the largest recorded crypto theft, and that the open questions are mechanism and recoverability, not whether the bureau believes its own attribution. The unsourced presidential-dispute reference deserves a placeholder, not a verdict. The week's other signals are consistent but modest: edge-device KEV entries with today's remediation deadlines, AI hallucination reaching courts, and AI infrastructure dealmaking whose terms remain unannounced. A reader should be firm on the attribution and the ETF outflow numbers, and humble on everything about how the funds were taken.

Independent Cross-Check — Kimi

A separate AI model (Kimi) independently read the same corpus. Agreement corroborates the desk's read; divergence flags a contested story. 1 China-sensitive story was withheld from it.

Certainty calls rate how settled the underlying facts are, not how the story is framed. Consensus: independent source types corroborate what happened. Contested: sources disagree on substance, or the story rests largely on one side’s reporting. Developing: thin or single-source coverage, or fast-moving and unconfirmed. Each call is the AI model’s own assessment of the day’s corpus.

Consensus 9 Contested 1 Developing 5

FBI attributes $6.3 billion Binance crypto theft to North Korean state-sponsored hackers Consensus

Multiple independent outlets (Security Affairs, Bleeping Computer, Decrypt, Futurism referencing presidential dispute) corroborate FBI attribution and approximate $6-6.4B figure; only framing/spin differs, with one source noting a presidential dispute of the claim.

Sam Altman makes unannounced visit to South Korea for chip partnership meetings Consensus

CNBC reports the visit with Samsung and SK Group; no contradictory accounts, basic fact of trip independently confirmable through host-country logistics.

Malaysia Airlines discloses customer data breach via Amadeus booking system Consensus

VnExpress and likely broader travel-industry coverage; airline's own disclosure provides factual anchor, no substantive dispute that incident occurred.

Seoul Viosys begins production of rare-earth-free white light emitter Consensus

Semiconductor Today reports with industry-specific sourcing; technical product claim is verifiable through supply-chain and patent records, no counter-claims evident.

About 80% of Korean judges have encountered AI-generated fake legal precedents in court cases Consensus

Korea Times cites Supreme Court survey data; specific institutional source provides factual substrate, though exact percentage depends on survey methodology.

Bitcoin ETFs saw ~$643 million in outflows following Binance hack Consensus

Decrypt cites specific fund-flow data; ETF flows are publicly trackable through fund administrators, making this independently verifiable.

Hackers breach Russia's Federal State Statistics Service (Rosstat), claim data on Russian losses in Ukraine Contested

Only Ukrinform (Ukrainian state-aligned outlet) reports the claim by Belarusian hacktivist groups; no independent corroboration from Rosstat or neutral outlets, and the hacktivists themselves have incentive to claim success regardless of actual data obtained.

OpenAI considering rebranding data centers as 'AI compute campuses' Developing

Single TechCrunch report based on unnamed sourcing or speculative framing; no second outlet corroborates this as a confirmed corporate decision rather than trial balloon.

Neolix and Sunway announce autonomous delivery partnership in Malaysia Developing

Only Antara News (Indonesian/Malaysian outlet) carries this announcement; corporate press release may underlie it, but no second independent source confirms the partnership terms or operational reality.

Humanoid robot traffic police deployed at 2026 Hangzhou Marathon Developing

Only Global Times (Chinese state media) reports this; no independent visual verification or second outlet coverage, and state media has incentive to highlight tech showcases.

President disputes FBI assertion that North Korea behind Binance hack Developing

Only Futurism brief mentions this dispute in a headline reference; no direct quote, named official, or second source confirms which president or the substance of the denial.

Japan, US, South Korea exploring Indonesia for small modular reactor deployment Developing

Only Japan Forward reports this trilateral interest; no corroboration from US, South Korean, or Indonesian outlets, and the story appears to stem from a single analytical/policy source.

Trigger.dev discloses security incident exposing customer environment variables Consensus

Company's own blog post is primary source, but security incidents of this nature are typically self-reported and verifiable by affected customers; Hacker News discussion provides community corroboration of incident reality.

Samsung announces partnership for Youth Olympic Games Dakar 2026 Consensus

Samsung's own newsroom release; corporate sponsorship announcements are self-documenting and verifiable through Olympic organizing committee records.

NASA seeks commercial industry assistance to extend Hubble Space Telescope operational life Consensus

The Verge reports based on NASA public statements; agency's open solicitation for private mission concepts is a matter of public record, no factual dispute.

Watch Next

Any primary statement from the White House or FBI clarifying the reported presidential dispute of the North Korea attribution
Binance or FBI technical disclosure on the intrusion vector and whether stolen funds are being frozen after the FBI's request to block linked transactions
Bitcoin ETF flow data for Monday and Tuesday to see whether the $643.2 million Friday outflow persists
Ivanti Connect Secure patching status for CVE-2026-111357, whose KEV remediation deadline is 2026-11-02, and CVE-2026-118527 due 2026-11-13
Further Amadeus-related airline disclosures following Malaysia Airlines, and any regulator response
Any official confirmation or terms from Samsung, SK, or OpenAI following Altman's Korea visit

Historical Power Lenses AI analysis

AI back-tests: the model applies each figure’s documented decision-making framework to today’s sources. These are not the figures’ own words, and the historical parallels come from the model’s general knowledge, not from the sources cited in this brief. Every lens, every cadence →

Genghis Khan 1206-1227

Genghis Khan's empire ran on information: his yam relay system moved intelligence faster than rivals could react, and he rewarded engineers and merchants whose skills he could absorb. North Korea's TraderTraitor operation mirrors the logic, treating theft as state revenue and treating foreign technical talent and platform access as the conquered resource. The FBI's request that firms block transactions is the equivalent of cutting the caravan routes that Khan's enemies relied on, and it works only if the private chokepoints cooperate.

Thomas Edison 1847-1931

Edison understood that the patent portfolio and the supplier relationship were as decisive as the invention itself, and he fought the current wars with both. Altman's unannounced visit to Samsung and SK is that playbook applied to memory capacity: lock the scarce input before the rival does. The 'AI compute campus' rebrand echoes Edison's habit of managing public acceptance of unpopular infrastructure, as when his Pearl Street electrification had to win over a city wary of wires overhead.

Cleopatra VII 69-30 BC

Cleopatra preserved Egypt's autonomy by trading on the one asset great powers needed, grain and money, and by choosing alliances with Caesar and then Antony to stay relevant. Samsung and SK sit in a comparable position: smaller than the superpowers around them, yet holding the scarce memory capacity that OpenAI and its rivals cannot do without. Her lesson is that leverage lasts only as long as no single patron can replace you, which is why the suppliers will want multiple buyers at the table.

Napoleon Bonaparte 1799-1815

Napoleon's Code Napoleon reformed law during wartime because he knew institutions outlast victories. The Korean judiciary's survey on fake AI precedents, and the AMA's liability framework, are early attempts at that kind of institutional response to a fast-moving disruption. His record also cautions that codes written under pressure tend to harden around the problem of the moment, and AI-verification rules risk the same fate.

Sources Cited

16 sources — show

Source types are read from each link’s address by fixed rules, not assigned by the model. Primary record marks what a government, court or company itself published; the other types are reporting or commentary about events. A link no rule identifies carries no type rather than a guess.

Lean labels: L Left · LC Lean-Left · C Center · RC Lean-Right · R Right · INTL International · GOV Government. INTL: Geography, not a left/right position: the prompts ask for a cross-section spanning left, right, center, international and government sources. GOV: A source type, not a political position. The model assigns it, and has applied it to state-affiliated media; the source-type label is derived separately from the URL. Lean codes on a brief's citations are assigned by the model that wrote the brief: an estimate, not an editorial rating. Where this site’s own outlet profile or domain rule gives a different label, that label is shown and the model’s follows in parentheses.

securityaffairs.com/202221/apt/fbi-blames-north-korea-for-6-billion-b… Center (C)
bleepingcomputer.com/news/security/fbi-links-63-billion-binance-crypt… Specialist (model: Center) News / analysis
decrypt.co/382309/bitcoin-etf-investors-exits-record-binance-hack Center (C)
futurism.com/future-society/investor-economy-collapse-ai-raoul-pal Center (C)
cnbc.com/2026/11/02/openai-sam-altman-samsung-sk-south-korea.html Center (C) News / analysis CNBC profile
techcrunch.com/2026/11/01/openai-might-rebrand-its-data-centers-as-ai… Specialist (model: Center) News / analysis TechCrunch profile
semiconductor-today.com/news_items/2026/oct/seoul-291026.shtml Center (C)
japan-forward.com/us-south-korea-indonesia-small-modular-reactors Center (C)
koreatimes.co.kr/southkorea/20261102/about-80-of-korean-judges-encoun… Center (C) News / analysis
fiercehealthcare.com/ai-and-machine-learning/whos-responsible-clinica… Center (C)
allenai.org/blog/novelty-bench Center (C)
trigger.dev/blog/security-incident-nov-2026 Center (C)
ukrinform.net/rubric-ato/4180280-hackers-breach-russias-statistics-ag… Center (C)
e.vnexpress.net/news/news/traffic/malaysia-airlines-becomes-latest-ca… Center (C)
federalregister.gov/documents/2026/11/02/2026-22385/adjusting-imports… Government (GOV) Government / official · primary record
bbc.com/hausa/articles/cvlydp1kk7j9o?at_medium=RSS&at_campaign=rss Center (C) News / analysis BBC News profile

Reference

Chart auto-generated from this brief's structured fields. See methodology for how the underlying data is collected.

Other desks

Intelligence Desk Markets Desk Defense & Security Desk Energy & Climate Desk Insurance Desk Health & Science Desk Culture & Society Desk Sports Desk World Desk Local Wire Politics Desk