Joining the dots

I am reproducing a Substack article here, relevant to technology applications and their Corporate owner politics of the present day:

Part 4: She Won. Or did she? A Deeper Look at Tripp Lite, Starlink, and the Air Gap Question.

Unpacking A Key PIece to the Election Integrity Puzzle

Michael D. Sellers

Jun 20READ IN APP

A number of commenters have pointed out that my analysis so far has not included the issue Tripp Lite (Eaton) UPS issues, which has been identified by This Will Hold, Smart Elections, and others as a key vulnerability in the 2024 (and potentially future) elections. I’ve done a deep dive into this. Following is my report.

At issue is a device called Tripp Lite (Eaton) UPS, with UPS referring to Uninterrupted Power Sourse. In simple terms, a UPS is like a safety net for electronics, providing emergency power and surge protection so devices don’t crash or get fried by power spikes. In election settings, UPS units are used to keep voting equipment running smoothly even if the power goes out. For example, ballot scanners (tabulators) often connect to UPS devices so they won’t shut down mid-count or lose memory if there’s a power flicker. Electronic poll books (the tablets or laptops that check voters in) and central vote tally servers may also rely on UPS backups or built-in batteries to ensure voting can continue uninterrupted. These backup systems are a part of election infrastructure.

Lately, these battery boxes are drawing scrutiny because a UPS isn’t just a battery; modern units can also be “smart” devices that connect to networks. That’s great for IT managers who want to monitor battery health or get an alert if power fails. However, in a voting context it raises a big question: Could a “smart” UPS inadvertently connect voting machines to outside networks or actors? In other words, could the very device that’s meant to protectelection equipment also introduce a hidden way to access it?

Tripp Lite, Eaton, and the Hidden Player in Election Infrastructure

Tripp Lite is a Chicago-based manufacturer of power strips, surge protectors, and UPS units. Tripp Lite’s hardware has quietly become embedded in our voting systems. Certification documents from major election vendors show that for years Tripp Lite devices were recommended components: election machine manuals list specific models of Tripp Lite surge protectors and UPS units as preferred or required accessories. Starting around 2018, ES&S (which serves roughly 60% of U.S. voters) began advising counties to plug certain high-speed ballot scanners (the DS450 and DS850 models) into Tripp Lite’s “SpikeCube” surge protector. Dominion Voting Systems, in its 2019 and 2020 documentation, actually named a particular Tripp Lite UPS as the primary backup power unit for its election management server (with a few pricier brands as alternatives). By 2024, Dominion’s updated guidelines still listed the Tripp Lite UPS first among five options for any critical equipment. Election officials don’t always have to buy the first-listed brand, but many likely do – it’s a safe choice to match what the certification tested. The result: in hundreds of counties, the voting machines counting mail-in ballots, the servers aggregating results, and perhaps other devices were all plugged into Tripp Lite power units.

Why is this drawing attention? Because of who owned Tripp Lite. For over 50 years, Tripp Lite was a private company led by a single individual – Barre Seid, a secretive Chicago electronics magnate. In 2021, Seid made headlines for an unprecedented act: he donated 100% of Tripp Lite’s stock (worth around $1.6 billion) to a nonprofit controlled by Leonard Leo, a powerful conservative activist. Leo’s nonprofit quickly sold Tripp Lite to the Irish-American conglomerate Eaton Corporation for about $1.65 billion. Effectively, the proceeds of this sale became a massive political war chest for conservative causes (the largest single donation of its kind in U.S. history. In short, the UPS boxes guarding our election machines were until recently owned by a man who poured the windfall into an organization dedicated to reshaping courts and elections to conservative ends. That alone is a striking intersection of money, politics, and critical infrastructure.

Eaton, the new owner, is a Fortune-500 power management company not typically associated with election controversy. They now market Tripp Lite’s products under the name “Tripp Lite by Eaton.” Yet observers note that since acquiring Tripp Lite, Eaton has formed alliances with figures and companies tied to the 2024 election narrative. In May 2024, Eaton announced a partnership with Palantir Technologies – the data analytics firm co-founded by Peter Thiel – to integrate advanced AI into Eaton’s operations. Palantir’s software is used for everything from military intelligence to supply-chain management, and Thiel, a prominent tech billionaire, is known for his support of former President Trump. The Eaton-Palantir collaboration was about enterprise AI and resource planning on paper but it raised eyebrows among election integrity researchers. Why would a power equipment company need Palantir’s cutting-edge (and sometimes Orwellian) data technology? Some critics saw a red flag: a Thiel-linked company now had its foot inside a corporation whose hardware touches many voting systems. Eaton also struck a partnership with Elon Musk’s Tesla in late 2024 (focused on home energy storage and “smart” electrical panels). On its face, that was about solar power integration, not elections. But when Musk and Thiel – two billionaires often aligned politically – both turn up in the orbit of a company that, through Tripp Lite, quietly underpins parts of our election infrastructure, it’s hard not to ask what’s going on.

To be clear, none of this proves any wrongdoing. It could be coincidence or the natural consequence of Eaton trying to innovate via partnerships. What it does show is that the power management layer of our voting process, once an afterthought, is now entangled with partisan billionaires and high-tech firms. And that’s why the next part of this story – about potential digital vulnerabilities – has gained traction. The people and money behind these “dumb” power boxes warrant scrutiny, but so does the technology inside them.

DEEPER LOOK with Michael Sellers is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

Upgrade to paid

Theory of the Case: Did “Smart” UPS Units Bridge the Air Gap?

One of the principles of election security is the “air gap” – keeping voting machines physically and digitally isolated from the internet or any external networks. Officials reassure the public that ballot scanners and tabulators are not online, cannot be hacked remotely, and therefore the count can’t be tampered with from afar. In the 2024 election, as claims of Starlink satellites and wi-fi hacks flew around social media, state and federal authorities repeatedly emphasized that voting machines simply weren’t connected to any network during voting. But what if an interloper found an indirect way in – not by plugging into the voting machine, but into something that’s plugged intothe voting machine?

This is where the Tripp Lite UPS devices come under the microscope. Many of these UPS models are not just batteries; they’re network-capable computers in their own right. Tripp Lite (now Eaton) sells optional add-in cards – essentially small circuit boards with a network port – that turn a UPS into a smart node on your network. For instance, the SNMPWEBCARD accessory allows administrators to connect a UPS to an Ethernet network and monitor or control it via web browser or SNMP (Simple Network Management Protocol). With a network card installed, a UPS can send alerts, log power events, perform self-tests, and even execute remote shutdowns or reboots of connected equipment. In an ordinary office, that’s a feature – you can ensure servers shut down safely or remotely restart a crashed device by power-cycling its outlet.

In a polling place or election office, however, a “smart” UPS could inadvertently act as a bridge between the supposedly isolated voting equipment and an outside network. How? Consider a typical setup: a ballot tabulator is connected to a UPS for power backup, often via a simple power cord and sometimes a USB cable (for power-status signaling). If that UPS also has an active network cable (for instance, to report its status to a central console or simply because it was left plugged into an IT network from testing), you now have a chain linking the voting machine to an external network. The voting machine trusts the UPS (it’s just supplying power, after all), and the UPS is talking to the wider world – that is the essence of an unintentional digital bridge across the air gap.

Such a bridge could be exploited in several ways. The most straightforward is through the network: a hacker who can reach the UPS’s network interface might manipulate it to affect the connected machine. For example, security researchers have shown it’s possible to hack power management devices – as far back as 2012, one team created a malicious power strip that covertly served as a network backdoor. UPS units themselves have had vulnerabilities (e.g., weak default SNMP credentials or outdated firmware) that attackers could use to gain a foothold. If an adversary took over a UPS, they might shut it off at a strategic moment, cutting power to a voting machine (causing chaos or data loss). More deviously, a compromised UPS with a data link to the machine (via a USB management cable) might attempt to alter data or install malware on that machine. This scenario is highly complex – it would likely require a tailor-made attack to jump from the UPS’s microcontroller to the voting machine’s system. But researchers note it’s not impossible, especially if insiders were involved in designing or tampering with the hardware in advance.

There’s also the realm of side-channel attacks – ways to send or receive data through unconventional means (like power signals). Astonishingly, studies have demonstrated that it’s possible to extract data from an “air-gapped” computer by analyzing minute fluctuations in its power consumption or grounding. In 2018, Israeli researchers showed they could exfiltrate data from a laptop through the power lines by injecting signals into the ground connection – essentially turning the power cable into a low-bandwidth data transmitter. In theory, a smart UPS or even a modified surge protector could facilitate such powerline communication. This is far from a plug-and-play hack – it borders on spycraft – but it’s publicly known to be feasible. And if it’s feasible, one has to consider whether a well-resourced adversary (say, hypothetically, an alliance of powerful interests with access to engineering talent) could have built a capability to siphon off vote data or inject false data using the very power infrastructure of the voting machines.

The core of this theory is admittedly complex and so far unproven. The idea that a UPS could serve as an election Trojan horse sounds like something out of a technothriller. However, it’s precisely the kind of multi-layered covert approach one might use to compromise a system that everyone assumes is offline and secure. It wouldn’t matter if the voting machines weren’t on the internet if someone found a different route in.

How Starlink’s Direct-to-Cell Satellites Could Widen the Backdoor

So where do Elon Musk’s satellites come into play? In Part II of this series, we explored how a new network of low-Earth-orbit satellites might have quietly set the stage for unprecedented connectivity in 2024. These aren’t the regular Starlink internet dishes that rural homes use, but rather Starlink’s “Direct to Cell” system – satellites that communicate directly with everyday mobile phones. Musk’s SpaceX began launching these specialized satellites (essentially “cell towers in space”) in 2023 and by the end of 2024 had deployed the first full shell of them.The goal is to eliminate mobile dead zones by allowing 4G phones to connect from anywhere via satellite. By late 2024, the Direct-to-Cell service was active (in beta) for simple text messaging in parts of the U.S., with hundreds of satellites able to relay signals where terrestrial cell service didn’t reach.

For our scenario, Starlink Direct-to-Cell could act as the ultimate getaway car for data – or the entry point for a remote attacker – bypassing traditional network infrastructure entirely. Suppose an attacker managed to compromise a UPS or surreptitiously install a tiny device alongside it (imagine a cellular modem no bigger than a flash drive, tucked inside a power strip or UPS casing). Normally, that cellular device might struggle to get signal out of a sealed voting center, or there might be no cell tower nearby. But with satellites overhead acting as cell towers, that device could beam out data from inside the building directly to the sky. It’s Bond-movie stuff, but the technology exists now, not in some distant future. In practical terms, this means that even if election officials did everything right – no Wi-Fi, no internet cables, strong physical security – a cleverly placed cellular-based exploit could still punch a hole in the air gap. Starlink’s system would just grease the wheels, ensuring that any such malicious device had connectivity anywhere on the planet, without relying on local telecom networks.

Consider an example: a compromised UPS unit could be quietly transmitting a trickle of information over its power connection or USB port, and an implanted module in it or attached to it could aggregate that data and send it via a satellite link. That could provide live updates of vote tallies before they’re officially reported, a potential boon to someone trying to get ahead of election results or even manipulate betting markets. (In fact, one theory has pointed to unusually prescient betting market movements on election night, suggesting that someone might have had early knowledge of precinct results – a tantalizing but unproven hint that data was leaking out.) Alternatively, the satellite link could allow an attacker to send commands into the UPS or connected system at just the right time – for instance, triggering those UPS units to reboot or alter power flow during the vote count, potentially causing errors.

It’s important to stress that Starlink’s Direct-to-Cell service was brand new in 2024 and limited in bandwidth. Early on it could handle texts and maybe low-speed data, not streaming video or massive files. But for something like transmitting vote totals (which are just numbers) or small packets of instructions, that bandwidth would be more than enough. Also, while SpaceX’s public aim was benevolent (texting from remote mountains, emergency response connectivity, etc.), any technology can be dual-use. If indeed an election manipulation operation was afoot, the convergence of Musk’s satellite network and Thiel-linked power devices in polling places is conspicuous. Starlink provided a possible covert communication channel that no county IT department would likely detect, since it operates outside the traditional internet and cell grid.

Again, we have to keep one foot on the ground: this is a hypothetical layering of vulnerabilities – if someone rigged some UPS units and managed to use Starlink to communicate, it could bridge an air gap that was assumed impenetrable. It’s a lot of “ifs.” But it’s also the kind of outside-the-box thinking that election security folks are paid to consider. And here we have the pieces on the chessboard: UPS devices in key positions, network-enabled and possibly overlooked, and a new satellite network blanket that can whisper to devices anywhere, courtesy of a billionaire who was openly rooting for a particular election outcome.

Thiel, Palantir, and Why Critics Are Concerned

Let’s circle back to Peter Thiel and Palantir’s connection to all this, because it speaks to motive and opportunity – the classic questions of any investigation. Peter Thiel is not just a tech investor; he’s a political actor. He supported Donald Trump in 2016 and has funded candidates and causes aligned with Trump’s agenda. Palantir, the firm Thiel co-founded, built its reputation on crunching big data for intelligence and defense – finding needles in haystacks of information. So when Eaton (the owner of Tripp Lite) deepened its partnership with Palantir in 2024, some observers did a double take. The partnership’s official purpose was to use Palantir’s Artificial Intelligence Platform (AIP) to streamline Eaton’s operations and data management. But skeptics wonder if there could be more to the story. Was Palantir’s software (and by extension Thiel) getting access to data about where Tripp Lite devices were deployed or how they were performing? Could Palantir’s vaunted analytics conceivably be turned toward analyzing election-related data flows if one were so inclined? These are speculative questions, and to be fair, there’s no public evidence Palantir had anything to do with elections in 2024. Eaton’s announcement didn’t mention voting at all, and Palantir’s work with them by all accounts stayed in the lane of logistics and enterprise AI.

Critics view Thiel’s proximity here as a possible red flag simply because of his track record. It might be guilt by association, but it’s an association worth noting. In the narrative laid out so far: Barre Seid (Tripp Lite’s owner) wanted to advance conservative causes, Leonard Leo (to whom Seid gave the company) is an architect of conservative power-building, Peter Thiel is a key funder of the new right wing, and Elon Musk was an increasingly vocal supporter of the Trump comeback effort. All of them, improbably, have connective tissue to this corner of technology that touches our election systems. From a purely factual standpoint: Seid’s donation is known. Leo’s sale to Eaton is known, Eaton-Palantir collaboration is known, and Musk’s Starlink D2C network went live in 2024. What’s speculative is the notion that these facts were coordinated parts of an election subversion plan. That remains unproven. Yet, given the stakes, critics argue we can’t afford to dismiss the overlap as mere coincidence without investigating further. The concern is less “smoking gun” and more “smoke in the room” – enough to justify a careful look.

Thiel’s Palantir, in particular, raises philosophical alarms for some because of what Palantir does best: integrate disparate data streams and uncover patterns. If one were trying to surreptitiously influence an election via tech means, having a Palantir-like capability to monitor networks, sift logs, or coordinate devices would be extremely powerful. There’s noevidence Palantir software was deployed in election systems in 2024, but the fact that Palantir was effectively on call with Eaton suggests a theoretical vector: maybe not direct manipulation, but perhaps assistance in covering tracks or optimizing an operation. Again – this is conjecture. We mention it to clarify why some analysts are uncomfortable seeing Thiel’s shadow near the election infrastructure space. The flipside is also true: if nothing nefarious happened, these connections could be entirely innocent business deals being misinterpreted by outsiders. Keeping an even hand, we must distinguish what we know from what we fear: we know a Tripp Lite sale funded Leo’s networkr; we knowEaton works with Palantir and Tesla; we know Musk and Thiel had clear preferences in 2024. We do not knowif any of them leveraged Tripp Lite devices or Starlink for illicit purposes.

Where the Investigation Stands

As of today, all of this – the UPS bridges, the satellite links, the Thiel connection – remains a theory of vulnerabilityrather than evidence of an actual attack. No conclusive proof has emerged that any Tripp Lite (Eaton) UPS units were exploited to tamper with votes in 2024. In fact, election officials and cybersecurity agencies have flatly denied that any such interference occurred. The Cybersecurity and Infrastructure Security Agency (CISA) stated after the election, “we have no evidence of any malicious activity that had a material impact on the security or integrity of our election infrastructure.”. Officials in key states went on record specifically to debunk the idea that Starlink or other outside networks could have altered vote totals.. In short, the known facts of the 2024 election outcome remain that Trump won through the certified vote counts, and no official body has found those counts were fraudulently changed.

The UPS theory, then, is essentially a set of unanswered questions and anomalies. We know these devices were widely used (by design) in the election. We know the people behind them had motives to favor one side. We know technical pathways could exist to exploit them. But we do not have an example of “here is a county where a UPS was hacked and votes were changed.” There is some circumstantial smoke – for instance, statistical oddities in vote reporting or the betting market clues mentioned earlier – but nothing that rises to the level of proof. If a sophisticated attack happened via this route, it was subtle enough to evade detection so far. It’s also possible that the vulnerability was real but wasn’t actually used – perhaps plans were made or access was prepared (the pieces put in place) but ultimately not deployed on Election Day.

Investigators (both official and independent) are in the early stages of looking at this. Some journalists and researchers have begun filing public records requests to counties, asking for inventories of their UPS devices, maintenance logs, or network configurations. A few counties have disclosed that they did use Starlink kits for emergency connectivity, but only for transmitting unofficial results after polls closed, and not connected to voting machines. Those disclosures align with normal contingency planning, not a hidden conspiracy. So far, nothing public has directly implicated Tripp Lite UPS units in any wrongdoing. The theory remains just that – a theory – albeit one built on plausible concerns.

It’s also worth noting that this entire line of inquiry sits at the intersection of cybersecurity and partisan intrigue, which means it’s very easy to get out over our skis. The goal here isn’t to declare that “the election was definitely hacked through the power supply” – it’s to examine whether we’ve overlooked a potential weak link. Experts generally agree that U.S. election systems have become more resilient since 2016, with paper ballot backups and rigorous audits. Any would-be attacker would have to navigate a gauntlet of safeguards. That doesn’t mean it’s impossible – just difficult. If an operation targeting UPS devices did occur, it would rank among the most complex cyber-physical feats in election history. At this point, the most responsible conclusion is that it’s an intriguing hypothesis that warrants further investigation, not an established fact. The burden of proof is high, as it should be when questioning the result of a democratic election.

What Can Be Done: Next Steps for Transparency and Security

Even as we await more evidence, there are concrete steps that can be taken to address the worries raised by this theory. In the spirit of ensuring our elections are secure (and feel secure to the electorate), here are some paths forward:

1. Full Inventory and Audit of Power Devices: Election jurisdictions should conduct and publish audits of all auxiliary equipment used in elections – not just the voting machines and software, but things like UPS units, surge protectors, networking gear, and routers. Knowing exactly which models were used where, and whether they had network capabilities enabled, is fundamental. For example, if a county used Tripp Lite Model XYZ UPS with an SNMP card on their central tally server, that should be documented. Investigators (or independent experts) can then examine those specific models for vulnerabilities. If any UPS firmware updates were applied around the election, that’s a potential red flag to scrutinize (was there a software patch that could have introduced a backdoor?). This kind of audit doesn’t imply guilt; it’s basic cyber hygiene. Some states already have robust asset tracking, but making that information public (or at least available to oversight bodies) would build confidence.

2. Freedom of Information Act (FOIA) Inquiries: Journalists and citizens can file FOIA or state public records requests for communications between election officials and vendors about UPS devices or Starlink. If, say, a county elections director had an email exchange with a vendor about installing a network card in a UPS, or troubleshooting a Starlink unit at a vote center, those communications could shed light on how these technologies were used. FOIA requests could also target any incident reports – for instance, did any precincts report strange errors with their UPS during voting? Was there any unexplained power event that almost went unnoticed? Such leads could be valuable. While FOIA can be slow, it’s a legal avenue to get answers and create an archive of facts.

3. Independent Forensic Reviews: In extreme cases, one might literally forensically examine a few of these UPS units – especially any that behaved oddly. This would involve pulling the device’s internal logs, checking the firmware for tampering, and maybe even performing a circuit analysis to ensure no extra components are present. It’s somewhat analogous to a voting machine forensic exam (which has been done in past controversies). Because a UPS is not typically considered sensitive election equipment, there may be less resistance to letting an independent technical team inspect one. If nothing else, it could either rule out certain attack vectors or, conversely, discover something like a covert cellular module in a device (however unlikely that may sound). Even just verifying that all network cards were properly configured (or ideally, absent) in election UPS units would be reassuring.

4. Vendor Transparency and Patches: Eaton (Tripp Lite’s parent) should be invited to participate in this conversation. If their equipment is being questioned, they have an interest in clearing the air. The company could proactively disclose any known vulnerabilities in their UPS firmware or SNMP cards and detail what security measures they have in place. They might, for example, clarify whether the UPS network cards can operate over cellular or if they strictly use wired ethernet. They could also assist counties in updating firmware or disabling any remote access features during elections. Ideally, vendors of all election-related tech (even indirectly related, like power systems) should adhere to higher transparency standards when their products are used in the voting process. This doesn’t mean giving away intellectual property – it means recognizing that democracy is a unique customer that requires unique assurances.

5. Policy and Procedural Changes: On the governmental side, this saga highlights that “air-gapped” needs a stricter definition in election regs. Officials might consider rules that any device connected to a voting machine (even if it’s “just a power supply”) must itself be free of network connectivity during the election. That could mean requiring UPS units with no network cards installed, or physically sealing those ports on Election Day. Procedures could be updated so that, for instance, if a UPS or power device fails, it’s reported and examined, not just swapped out. Another policy idea is mandating post-election audits not only of ballots but of equipment logs – checking if, say, any unexpected IP addresses contacted a piece of equipment. These steps can harden an already secure system by shoring up an obscure corner of it.

6. Ongoing Investigation and Open Minds: Finally, we as investigators (in the broad sense, including the concerned public) should follow the evidence wherever it leads – even if it debunks our worst fears. That means if forensic audits and FOIAs come back showing nothing but mundane explanations, we have to be ready to accept that. Conversely, if they uncover irregularities, we must pursue those rigorously but responsibly. The tone of this series has been to question, not to accuse. That must continue. The worst outcome would be to jump to conclusions or let partisans hijack these findings to spread baseless allegations. The best outcome is we either confirm a vulnerability and fix it or confidently rule it out and reassure the public.

In conclusion, the story of Tripp Lite UPS devices and Starlink satellites in the 2024 election is a reminder that election security is a constantly evolving challenge. Threats that sounded far-fetched yesterday can become very real tomorrow. Who would have thought a decade ago that satellites talking to voting machines would be a serious topic? Or that we’d be analyzing the politics of an unassuming power supply company? Yet here we are. The good news is that awareness is the first step toward resilience. By shining light on these shadowy corners – the backup batteries, the network cards, the satellite links – we empower election officials and the public to ask the right questions and demand solid answers. Our democracy’s strength lies not just in robust systems, but in the transparency and trust that come from scrutinizing those systems. If something did happen in 2024, it’s important to uncover the truth and deal with it. If it didn’t happen but could have happened, then it’s important to learn from that and apply the knowledge going forward, to restore voter trust going into 2026 and 2028. That’ a tall order, under the circumstances, but it’s essential to work toward that objective.

DEEPER LOOK with Michael Sellers is a reader-supported publication. This is a lot of work! 😉 Please support that work if you can!

Upgrade to paid

You’re currently a free subscriber to DEEPER LOOK with Michael Sellers. For the full experience, upgrade your subscription.

Upgrade to paidLikeCommentRestack

© 2025 Michael D. Sellers
3302 S Grand Ave, Los Angeles, CA 90007
Unsubscribe

Start writing
Unknown's avatar

About borderslynn

Retired, living in the Scottish Borders after living most of my life in cities in England. I can now indulge my interest in all aspects of living close to nature in a wild landscape. I live on what was once the Iapetus Ocean which took millions of years to travel from the Southern Hemisphere to here in the Northern Hemisphere. That set me thinking and questioning and seeking answers. In 1998 I co-wrote Millennium Countdown (US)/ A Business Guide to the Year 2000 (UK) see https://www.abebooks.co.uk/products/isbn/9780749427917
This entry was posted in anthropocene and tagged , , , , , , . Bookmark the permalink.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.