A common complaint about the rise of commercial AI services is that they are power-hungry and thus damage the environment. If this concerns you then [Squeezlabs] has the solution, in the form of an AI powered by a handcrank.

The guts of the system is a Raspberry Pi 5 running llama.cpp and appropriate speech conversions, but it and the Large Language Model (LLM) side are not the most interesting part of this system. The power comes from a hand crank charger of the type you’ll see for sale on the likes of AliExpress, designed for USB charging. That in itself is not enough to power the Pi though, as upticks in the processing can cause brownouts that crash the machine. Thus there’s a custom-made capacitor board to take up the strain, and even with that the handle resistance varies significantly depending on the computing load.

We can see that this is not the ideal way to experience an LLM, but maybe that’s not the point. It does however point towards a future in which the power demands of processing decrease and less effort is required. Meanwhile, this is by no means the first hand cranked project we’ve seen.

hackaday.com/2026/06/11/ai-the…

[Dennis] is on YouTube with his channel “Made By Dennis,” but for the record he is a maker, not a V-tuber. On the other hand, his latest project– creating a profesisonal-level tracking rig with DIY IR cameras and a whole lot of moxie–does mean he’s now equipped to make the move to the prestigious, high-status world of pretending to be an anime girl.

That is of course not why he did it. Like most projects around here, the motivation was more a case of “I wonder if I can…”– in this case [Dennis] wondered what it would take for him to pull off the same sort of optical motion capture, or MoCap, that is used in Hollywood studios. Optical mocap has the advantage of being very precise, able to track things at high speeds, and not being in any way limited to the human form like the slew of AI-assisted methods hitting the market right now. The disatvantage is that you need to place markers on any part of your subject you want tracked, film them from all angles, and process a whole lot of pixels. In [Dennis]’s case, it ended up being about four billion. Keeping in mind that actually locating those points in 3D space is dependent on knowing exactly where your cameras are: if you want sub-millimeter precision, your cameras need to be fixed with sub-millimeter tolerance. It’s a big project, hence a long video, which is embedded below.

The DIY cameras use a AR0234 MIPI camera on a custom PCB with M12 lenses and IR filters. To improve the signal-to-noise ratio on optical MoCap, it’s standard to use near-IR light. The camera boards, as you might expect given the MIPI interface, hook into Raspberry Pi compute modules– the cheapest CM4 should work, though he’s using CM5s. The compute modules sit on custom boards that provide PoE, and some other niceties– like a small microcontroller driven by the pulse-per-second pin to help trigger the cameras in sync.

Each camera gets a ring light of near-IR LEDs that pulse at 160 W, which would be way more than PoE is specced to provide, but since the LEDs are only on when the camera is taking a frame, the average power is well within allowable limits. With 16 cameras each having their own ring light, that’s a lot of near-IR photons. Don’t forget your safety squints!

Rather than process the images with OpenCV, he has his own custom solution optimized for this use-case that [Dennis] reports is 300x faster. Luckily, he’s put his implementation on GitHub, along with the rest of the project. Even if you don’t have any v-tubing ambitions, this project is very impressive and worth checking out in its entirety.

Optical MoCap isn’t the only game in town, of course. If you want to do this cheap and easy, you can strap a bunch of IMU sensors to yourself– just don’t expect the same precision.

Thanks to [Dennis] for the tip!

youtube.com/embed/kYVqL_DqBis?…

hackaday.com/2026/06/10/proces…

A buckminsterfullerene, also known as a buckyball, is typically a fullerene consisting of sixty carbon atoms (C₆₀) arranged in a way that resembles a football-like sphere. Extending this arrangement to other types of atoms has until now however proven as illusive as finding non-carbon-based lifeforms. In a paper by [Hyun Wook Choi] et al. and published in Chemical Science the discovery of boron buckyballs is detailed. There is also a soft-paywalled article in the Chemical & Engineering News magazine for a higher-level perspective.

The discovered boron-based buckyball ups the number of atoms to eighty, forming B₈₀ (boron fullerite) with a slightly larger diameter than C₆₀ at 0.85 nm versus 0.71 nm. Perhaps more interesting are the claims by the authors that boron fullerite may have more practical applications than its carbon-based cousin, mostly due to it being predicted to be a semiconductor with an 0.8 eV energy gap and better electron acceptance that provides interesting doping prospects.

Producing these boron structures used laser vaporization with a helium carrier gas that was seeded with argon to increase cooling efficiency. Inside this boron cluster the reported structures were then discovered and characterized as described in the paper.

Obviously, going from a fascinating laboratory discovery to bulk production won’t be easy, and the predicted properties of boron fullerite may turn out to be incomplete or have a dark side that we aren’t aware of. Regardless, they’re bound to be more useful at least than the carbon version that’s remained mostly a curiosity despite many years of research.

hackaday.com/2026/06/10/introd…

No, we’re not talking about cultural appropriation of Japan’s most famous form of short poem–this is the other Haiku, the open-source descendant of BeOS, which now has a fully-native meshcore chat client called Sestriere, thanks to the efforts of one [Atomozero]. Of course you’ll need a LoRa radio to act as a modem, but anything that speaks USB serial– which is any of the ESP32-based offerings on the market–should work.

This is interesting in that we don’t see many desktop applications leveraging LoRa networks– meshtastic or meshcore– so for one to appear for the relatively-obscure BeOS derivative is just neat. It’s also a nice peice of work: the chat window is full featured, organizing your contacts, and communicating not just with text but emojis and reaction GIFs. GIFs seem a bit extravagant for LoRa bandwith, but apparently it works. There are also Codec2-based voice messages, another thing that we didn’t expect to see over LoRa, since most ‘chat’ projects restrict themselves to text messaging.
The chat window. One nice thing about Haiku APIs is that look-and-feel isn’t in question.
The software will also map all the nodes with which you are in contact, both diagrammatically and geographically, overlaid on OpenStreetMap tiles. The network map conveniently colour-codes your contacts by the link quality, but what’s even more interesting is the WireShark-inspired packet sniffer built into the software to let you keep a really close eye on traffic on the mesh network.

Neither Haiku or MeshCore are to everyone’s tastes, but as an OS it is a worthy daily driver, even if you have to jump through some hoops to install it if you have a UEFI-only system.

If you need more range, try a Yagi.

hackaday.com/2026/06/10/meshco…

As straightforward as the concept of taking battery packs out of an old electric or hybrid car and reusing them for home power storage sounds, this thought process skips a few essential steps. As argued by [Ed] in a recent video based on his own experiences with high-voltage Nissan Leaf batteries in a home PV system, the main problem is that you’re taking a battery out of a larger system including a lot of the management hardware and software.

The referenced Battery Emulator project is an open source effort to create a suitable interface between these EV batteries, with the mentioned Nissan Leaf being just one example in the project Wiki, with the connection scheme shown in the top image. It’s also noted that the Leaf battery BMS is not designed to operate continuously, so they need to be restarted every day or so lest they become too inaccurate.

These and other things are all solid reasons why you have to be absolutely certain that you want to integrate these high-voltage battery packs into your 12 – 48V low-voltage DC system. You’re after all assuming all the responsibility of setting up a system that’s both safe and reliable, so having a good read through something like the Battery Emulator Wiki and sourcing first-hand experiences from the folk in this community would be a very wise first step.

youtube.com/embed/zcyRrmwDoZg?…

hackaday.com/2026/06/10/safely…

Quindi grazie Daniele, la notizia è questa: Trump e Sam Altman, assieme nientepopodimeno che a Bernie Sanders, stanno lavorando su un'idea meravigliosa: una partecipazione pubblica al capitale delle aziende della cosiddetta "Intelligenza Artificiale" (Fortune).

TL;DR: Che gli dèi ci proteggano.

dk.dataknightmare.eu/dkx10x35-…

dataKnightmare Archive

2026-06-10 18:40:05

DK 10x35 - Partecipazione alle perdite

Ascolta l'episodio su Spreaker.com

La puntata di oggi parte da una segnalazione dell'infaticabile Daniele, che non permette mai che io vada in ipomadonio, che ormai c'ho un'età.

Quindi grazie Daniele, la notizia è questa: Trump e Sam Altman, assieme nientepopodimeno che a Bernie Sanders, stanno lavorando su un'idea meravigliosa: una partecipazione pubblica al capitale delle aziende della cosiddetta "Intelligenza Artificiale" (Fortune).

TL;DR: Che gli dèi ci proteggano.

Sigla.

Bernie Sanders, l'arzillo senatore socialdemocratico del Vermont, ha lanciato un'idea meravigliosa, talmente meravigliosa che raccoglie il consenso di Sam Altman e di Trump, perfino.
Un'idea così, rischiamo di vederla realizzata.

Usiamo le parole dello stesso Sanders, sul New York Times:

L'intelligenza artificiale sarà quasi certamente la tecnologia più rivoluzionaria nella storia del mondo. Influirà profondamente sulla vita di ogni uomo, donna e bambino nel nostro Paese. Porterà (e sta già portando) cambiamenti inimmaginabili alla nostra economia, alla nostra democrazia, al nostro benessere emotivo, al nostro ambiente e al modo in cui educhiamo e cresciamo i nostri figli. Inoltre, esiste il timore molto concreto che, man mano che l’intelligenza artificiale diventerà più intelligente degli esseri umani, possa alla fine funzionare in modo indipendente, con conseguenze potenzialmente catastrofiche.
(New York Times, 01 Giugno 2026)

Ora. Questa non è un'opinione: a parte quel "quasi certamente" che porta un sacco di peso, è la proposizione di vendita di ogni azienda della cosiddetta "Intelligenza Artificiale", qualcosa su cui negli ultimi cinque anni è stata dispiegata la più fenomenale operazione di marketing e propaganda della storia. L'idea che tutto cambierà perché ora ci sono i modelli linguistici è, nel migliore dei casi, ancora tutta da dimostrare.

Secondo. I meravigliosi guadagni sono del tutto immaginari, e fanno anch'essi parte della proposizione di vendita. Al momento in cui scrivo, la cosiddetta "industria" (si sentono le virgolette?) della cosiddetta "Intelligenza Artificiale", cioè per essere precisi dei modelli linguistici, meno l'unico fornitore, NVIDIA, ha speso 1200 miliardi di dollari, a fronte di 135 miliardi di introiti, e offre ogni livello di abbonamento in perdita. In altre parole, i "cambiamenti inimmaginabili" partono con 1065 miliardi di dollari di perdite.

Terzo. I pericoli della "IA che funzionerà in modo indipendente" sono la manovra di marketing con cui gli AI bro vogliono farsi grossi: è chiaro che se qualcosa minaccia l'umanità intera è una cosa seria e importante. Purtroppo, i modelli linguistici non sono né l'una né l'altra cosa, e le minacce esistenziali esistono soltanto nei trip sotto chimica degli AI bro. Non so cosa si fumino, ma a giudicare dagli effetti, è comprovabilmente roba troppo forte per loro.

Andiamo avanti con le parole di Sanders:

Poiché l'intelligenza artificiale si fonda sul sapere collettivo dell'umanità, la ricchezza che essa genera deve andare a beneficio dell'umanità. Non solo del signor Musk, del signor Altman, di Dario Amodei e di altri magnati le cui aziende sono posizionate per dominare il settore. Non solo dei venture capitalist della Silicon Valley o dei gestori finanziari di Wall Street che, senza dubbio, vedono l'intelligenza artificiale come la prossima grande macchina per la creazione di ricchezza.
(New York Times, 01 Giugno 2026)

OK. L'idea che i modelli linguistici siano stati costruiti acquisendo forzatamente il materiale di chiunque, senza un centesimo di negoziazione o di diritti, è almeno finalmente un fatto. Si è trattato di una pura e semplice estrazione di ricchezza di tipo coloniale.

Ma il fatto che siccome i modelli linguistici sono stati costruiti sulla somma complessiva delle conoscenze umane, allora gli statunitensi hanno il diritto al 50%, è un'idea altrettanto coloniale. Chi se ne stracatrafotte se al mondo ci sono sei miliardi di utenti Internet e otto miliardi di persone. L'importante è che l'umanità rappresentata da trecento milioni di statunitensi ne tragga profitto.

pausa

Dopo una vita in politica, Sanders non sembra avere uno stagista, un consigliere, un amico, come direbbe Luca Bizzarri, capace di ispirargli un attimo di riflessione, o almeno di dirgli che ripetere a pappagallo i lanci stampa delle multinazionali non è una buona base per una politica, men che meno socialdemocratica.

Sia come sia, Sanders continua senza nemmeno prendere fiato.

Ecco perché presenterò a breve l’American AI Sovereign Wealth Fund Act (Atto per il Fondo Sovrano sull'Intelligenza Artificiale). Questa legge, se approvata, consentirà ai cittadini americani di acquisire una partecipazione diretta nelle maggiori aziende di intelligenza artificiale del nostro Paese. In che modo? Verrà istituito un fondo sovrano attraverso un’imposta una tantum del 50 per cento; pagata non con i profitti di OpenAI, Anthropic, xAI e altre aziende, ma con qualcosa di ben più prezioso: le azioni.
(New York Times, 01 Giugno 2026)

L'idea di Sanders si chiama nazionalizzazione. In tempi normali, Sanders avrebbe già avuto un tragico incidente d'auto o un improvviso infarto prima che a Langley fosse l'ora dell'aperitivo. Ma non viviamo in tempi normali.

E infatti, non solo Sanders è vivo e vegeto, ma Sam Altman ha richiesto e ottenuto un incontro a due per parlare dell'idea, e Trump stesso, a bordo dell'Air Force One, ha detto ai cronisti che l'idea è sul tavolo e che riceverà a breve i vari Musk, Altman e Amodei.

Il cronista della Associated Press l'ha messa lì piano:

"C'è qualcosa di molto interessante nell'idea, diventa quasi una compartecipazione con il pubblico Americano" ha detto venerdì Trump, un Repubblicano.
Associated Press, 6 Giugno 2026

Quando i cronisti gli hanno fatto notare che Sanders si definisce socialdemocratico, che per inciso negli Stati Uniti è eresia pura, Trump ha ribattuto che la propria base e quella di Sanders, sulle questioni di economia non sono poi così distanti.

Grande è la confusione sotto il cielo.

La cosa divertente è che l'unica timida obiezione l'ha sollevata Altman, che ha detto di non essere proprio d'accordo sulla percentuale del 50%. Che tenero.

Vediamo di ragionare un po'.

pausa

Sono mesi che i vari Altman, Musk, Amodei si alternano in una settimana-tipo:

• il lunedì uno a caso dice che non è sicuro che il proprio modello linguistico non sia diventato senziente, e se non lo è adesso, di sicuro lo diventa entro uno o due anni;
• il martedì un altro ribatte che c'è la possibilità tangibile che la cosiddetta "Intelligenza Artificiale" annienti l'umanità, ma nel frattempo per tre o cinque anni gli Stati Uniti avranno aziende all'avanguardia;
• mercoledì il terzo dice che il secondo esagera ma che certamente l'industria della "Intelligenza Artificiale" è troppo importante e produrrà troppa ricchezza perché lo Stato stesso non se ne occupi in prima persona;
• giovedì gnocchi;
• venerdì Jensen Huang annuncia una nuova scheda dedicata che consuma solo un Megawatt invece dei due del modello precedente, così i data center possono comprare il doppio delle schede a parità di consumi.

Questa manfrina sta andando avanti da almeno un anno. Uno potrebbe pensare che a lungo andare qualcuno si accorga che ripetono sempre la stessa storia spostando in là i paletti temporali, e invece no. Apparentemente Sanders si beve tutta la storia e gli piace talmente da pensare di farci una legge ad hoc.

Ora, gli Stati Uniti sono un Paese dove se parli di sanità pubblica ti danno del comunista.

Arriva Sanders, propone di nazionalizzare il 50% dei giganti della cosiddetta "Intelligenza Artificiale" e non solo nessuno fiata, ma uno dei suddetti giganti e il presidente più intrallazzato business-friendly che si sia mai visto si dicono sostanzialmente d'accordo.

Sono solo io a pensare che i conti non tornano?

Che Trump possa essere d'accordo lo spiego col fatto che per lui "l'État c'est moi", quindi per Trump una partecipazione pubblica è una partecipazione di Trump.

Che Altman possa essere d'accordo lo spiego ancora più semplicemente: OpenAI ha più di 500 miliardi di dollari di debiti, e Sanders si sta offrendo di accollarne metà al popolo americano in cambio di una partecipazione nell'età dell'oro che la cosiddetta "Intelligenza Artificiale" promette entro cinque anni da cinque anni a questa parte.

Altman, ovviamente è un venditore di fumo ma non è scemo: e infatti ha detto che dell'idea di Sanders non gli piace solo la percentuale, perché il 50% vorrebbe dire che Altman non potrebbe più fare il bello e il cattivo tempo.

Altman è quello che fra le tante cose ha detto che openAI non può essere considerata un'azienda normale perché, parole sue, "siamo qui per il futuro glorioso".

E Altman è stato anche silurato per la quantità di balle che raccontava al proprio consiglio di amministrazione, prima che i soldi di Microsoft facessero tornare lui e dimettere tutti gli altri.

Altman sa benissimo che, anche con una partecipazione minoritaria, avere lo Zio Sam dalla propria parte significherebbe fiducia degli investitori al massimo e linee di credito sempre aperte, tanto paga Pantalone.

Continuo invece a non capire cosa ci trovi Sanders di geniale in quest'idea.

Per il contribuente, una partecipazione dello Stato nell'industria della cosiddetta "Intelligenza Artificiale" significa accollarsi, oggi, il 50% di 500 miliardi di debiti. Certo, poi nel futuro glorioso l'Intelligenza Artificiale ci renderà tutti ricchi (e pure biondi, giovani e con una minchia tanta), ma i miliardi di debiti sono qui oggi, ai guadagni (e al resto) ci crederemo quando li vedremo.

A me, lo sapete, la mitologica "mano invisibile del mercato", sembra sempre una favola per gli ingenui. Ma se come dicono i fan il mercato riesce benissimo a separare gli stupidi dal loro denaro, allora per questa volta opterei per un laissez faire in puro stile liberista. Niente partecipazioni agli utili immaginari, niente sostegni ai debiti reali, soprattutto niente salvataggi quando i creditori presenteranno le cambiali per cinquecento miliardi.

Lasciamo che i patetici buffoni che si atteggiano a grandi innovatori facciano i conti con quella libertà di impresa (che significa libertà di fallire) che sono così rapidi a invocare per le imprese degli altri.

Ricordate i cigolii di un paio di episodi fa? L'offerta pubblica di acquisto di SpaceX cigola già rumorosamente, e a cavallo del weekend pure OpenAI presenta la propria. Ma se gli investitori hanno già dei dubbi sull'offerta di Musk, che è un cialtrone, ma sta sul palco come nessun altro, l'aria non è buona.

Sanders dovrebbe mettersi il cuore in pace. Se vuole nazionalizzare qualcosa, provasse con Google Cloud Platform e AWS, che sono ormai a tutti gli effetti infrastruttura di interesse pubblico, e lasciasse la cosiddetta "Intelligenza Artificiale" ai sogni di un branco di bolsi quarantacinquantenni con la visione di un adolescente negli anni '70.

Bernie Sanders and Sam Altman's private one-hour meeting about the public ownership of AI | Fortune

Trump told reporters that executives from leading AI companies will visit the White House, “probably next week,” to discuss a potential "partnership."

^{Joey Cappelletti (Fortune)}

This week Jonathan chats with Alexander Neumann about Restic, a particularly compelling backup and restore solution written in Go. Why did the world need one more backup program? And what’s Alexander’s personal take on transitioning from programmer to maintainer? Watch to find out!

• restic.net/
• github.com/restic/restic
• fosstodon.org/@restic

youtube.com/embed/BSIV52iiWSU?…

Did you know you can watch the live recording of the show right on our YouTube Channel? Have someone you’d like us to interview? Let us know, or have the guest contact us! Take a look at the schedule here.

play.libsyn.com/embed/episode/…

Direct Download in DRM-free MP3.

If you’d rather read along, here’s the transcript for this week’s episode.

Places to follow the FLOSS Weekly Podcast:

• Spotify
• RSS

Theme music: “Newer Wave” Kevin MacLeod (incompetech.com)

Licensed under Creative Commons: By Attribution 4.0 License

hackaday.com/2026/06/10/floss-…

If you look around your environment, you can probably pick off quite a few things that you’ve made, at least if you’ve been at this a while. You probably aren’t reading this from the bottom of a body of water though, which means you lack the incredible confidence of submarine builder [Hank Pronk]. Not only is he building himself a capable-looking diesel-electric submarine over on YouTube, he’s even DIYing CO2 scrubbers for it! Yeah, that’s a man who believes in himself.

Luckily [Hank] is not anywhere near the Caribbean, so needn’t worry about being misidentified as a narco-sub, but he still has to be concerned about his oxygen supply when tooling around beneath the local lakes. Perhaps more important than the oxygen supply in a sub is the build up of CO2. It doesn’t matter how many oxygen tanks you bring down with you if you can’t scrub CO2 out of the air to make room for it. Just like the Apollo missions, he’s using a chemical adsorbent to take carbon dioxide out of the air — and just like Apollo 13, he’s switching from square to round.

Or, rather, from a rather rectangular commercial model to a DIY little round unit. That’s because he doesn’t need the big scrubber in this sub: being diesel-powered, he expects to spend a lot of time at snorkel depth, where both the pilot and the engines can get clean air through the tube. Dives are expected to be short, and in that use case, too big of a CO2 scrubber is really a waste. If for some reason he gets stuck on the bottom, well, the lake isn’t that deep. He can swim to surface, and has a detailed bailout plan. If he wants to stay under overnight to avoid bailing at night, he’s carrying enough extra adsorbent for that.

There’s a reason almost every submarine we’ve featured on this site over the years is an ROV. It’s not that a homemade submarine is automatically a death trap, but you sure do have to be confident in your design.

youtube.com/embed/H4Ek-IXaE10?…

hackaday.com/2026/06/10/diy-co…

When the Artemis lunar program was first conceived, the third mission would have seen astronauts step foot on the Moon for the first time since Apollo 17 in 1972. But as hard as getting into space is, a sojourn to our nearest celestial neighbor is even more mindbogglingly complex, and so earlier this year it was announced that actually landing on the Moon would be pushed out to the fourth mission.

In turn Artemis III would take a page out of the Apollo 9 playbook and test out rendezvous and docking procedures with commercial landers while operating in the relative safety of low Earth orbit. Moving the target date for the landing a few years down the road gave all involved parties a little more breathing room, but it also provided a valuable opportunity to gain insight into the performance of the vehicles and systems ahead of the critical moment. In the original timeline, the first time Orion would attempt to dock with the lander would have been just before descending to the lunar surface — leaving precious little time to troubleshoot should anything go wrong.

Yesterday NASA held a press conference to update the public on their progress towards the planned 2027 launch of Artemis III, which included the long-awaited announcement of the crew that will kick the tires on the next-generation lunar landers being developed by SpaceX and Blue Origin

Meet the Artemis III Crew

Randy Bresnik
Commander

A graduate of the Naval Fighter Weapons School (TOPGUN) and former F/A-18 Test Pilot, United States Marine Corps Colonel Randy Bresnik served as Mission Specialist aboard the Space Shuttle on STS-129 and Commander of the International Space Station during Expedition 53. He has logged more than 7,000 hours at the controls of nearly 100 types of aircraft, 3,600+ hours aboard spacecraft, and 32+ hours of spacewalk time between five extravehicular activities (EVAs).

Frank Rubio
Mission Specialist

United States Army Colonel Frank Rubio holds a Doctorate of Medicine and logged over 1,100 hours as a UH-60 Black Hawk helicopter pilot, with more than 600 hours of that time under combat conditions in Bosnia, Afghanistan, and Iraq. In 2022 he flew to the International Space Station aboard the Soyuz MS-22 on what was planned as a six month mission. But due to damage to the spacecraft, he ended up remaining on Station for 371 days, setting a new record for the longest spaceflight by an American astronaut.

Luca Parmitano
Pilot

European Space Agency (ESA) astronaut Luca Parmitano is a Colonel and Test Pilot in the Italian Air Force with 2,000+ hours of flying time on over 40 types of aircraft. He served as Flight Engineer on the International Space Station during Expedition 36/37 in 2013, during which time he became the first Italian to conduct an EVA. He successfully navigated a highly dangerous situation during his second EVA when a spacesuit malfunction caused his helmet to fill with water. He returned to the ISS in 2019 as part of Expedition 60/61, bringing his total time in space to just under 367 days.

Andre Douglas
Mission Specialist

Coast Guard Reserve officer Andre Douglas holds a Bachelor of Science degree in Mechanical Engineering, Master’s degrees in Naval Architecture, Marine Engineering, Electrical Engineering, and Computer Engineering, as well as a Doctoral Degree in Systems Engineering. During his time at the Johns Hopkins University Applied Physics Laboratory, he assisted in the development of NASA’s Double Asteroid Redirection Test (DART) mission and Japan’s Martian Moons eXploration spacecraft. He completed his astronaut training in 2024, and although he served as a backup crew member for Artemis II, this will be his first spaceflight.

One Mission, Three Launches

Although astronauts are by their nature the best of the best, the collected experience and knowledge of the Artemis III crew is truly incredible — and for good reason. This flight will be one of the most challenging and technically complex operations ever conducted in space, perhaps second only to the Apollo Moon landings themselves. In the most ambitious version of the plan, three spacecraft launched by three different booster rockets will conduct a carefully choreographed operation over the course of two weeks.

To start the first phase of Artemis III, Blue Origin will use one of their New Glenn rockets to carry the Blue Moon MK2 lander into low Earth orbit. The lander is designed to spend up to 90 days in space, which will give NASA a comfortable window of opportunity to get their Space Launch System rocket and Orion spacecraft ready for liftoff. After launch Orion will rendezvous and dock with the lander, and the crew will spend the next two days performing various tests and demonstrations. If everything goes well, they will ultimately enter the lander itself and don prototypes of the spacesuits that Axiom Space is developing for the Artemis IV crew to wear on the lunar surface.
Orion docking with Blue Moon MK2
Meanwhile, SpaceX will be preparing a modified version of their Starship V3 spacecraft for liftoff atop the Super Heavy booster. Once the Orion spacecraft is undocked and clear of the Blue Moon MK2, the prototype Starship Human Landing System (HLS) will launch and meet the capsule in orbit. According to SpaceX representatives, the vehicle itself won’t be too far removed from the version that completed a test flight back in May. Compared to Blue Origin’s lander, which will feature a boilerplate cabin design and functional life support systems, the Artemis III crew won’t be able to enter this early version of HLS.

Likely in expectation that comparisons would be made between the apparent capabilities of the two landers, SpaceX Vice President of Space Operations Jessica Jensen pointed out that many of the systems that will be used in Starship HLS such as the life support and avionics are derived from the flight-proven hardware used on the Crew Dragon — with some components such as the docking system being effectively identical. From the perspective of SpaceX, it’s more important to focus on testing the new hardware and procedures being developed specifically for the Moon.

Given that astronauts will not be able to enter the Starship HLS prototype, it’s expected the crew will spend significantly less time docked to it. After conducting some maneuvers to see how the two vehicles handle in relation to each other, the Orion will depart orbit and head for a splashdown in the Pacific.

Setting Course For Artemis IV

Although the press conference was about the upcoming mission, Jensen did give some brief details on how SpaceX and NASA are working together to refine the procedures for Artemis IV in 2028.
Artemis IV will now use a trajectory similar to the Apollo missions.
Back when Artemis III was set to touch down on the lunar surface, the plan was for Starship HLS to first enter into a relatively uncommon Near-Rectilinear Halo Orbit (NRHO) around the Moon, where it would eventually be met by the Orion capsule. However this was largely predicated on the idea that the Lunar Gateway Station would also be in NRHO. Now that the construction of Gateway has been abandoned, there’s no reason to rendezvous in that particular orbit.

Instead Orion will now dock with Starship HLS in low Earth orbit, just like it will on Artemis III. From there, Starship will use its own engines to perform the critical trans-lunar injection burn and put both craft on course towards the Moon.

This approach is not only easier to execute, but will require less propellant and therefore fewer refueling flights — directly addressing a common criticism leveled against the Artemis architecture.

High Risk, High Reward

Calling Artemis III ambitious would be an understatement. A mission involving a trio of spacecraft and their respective launch vehicles, two of which being early prototypes, has never been attempted in the history of spaceflight. Getting just one vehicle off the ground is a challenge in itself, and although experienced gained over the decades thanks to the International Space Station has made the subsequent rendezvous between two craft relatively routine, doing it twice during the same mission adds a whole new dimension.

Even the most ardent space fan has to admit it’s exceptionally difficult to believe that the involved parties can put such a bold plan into action in the next ~18 months, especially given the recent New Glenn explosion that has left Blue Origin’s launchpad in shambles. But it will certainly be exciting to see them try.

hackaday.com/2026/06/10/nasa-a…