From about 2017, individuals began to truly connect with the initial wave of companion robots. These devices had personality, moved around, joked, and answered when you spoke to them. Most early companion robots, however, were still limited by simple voice-command interactions and narrow functionality. Once the novelty wore off, many ended up sitting unused on shelves. As some of those companies went out of business and turned off their servers, many owners likened it to losing a pet.
What Ollobot describes as “gentle intelligence” is a useful way to think about where the serious work in this category is going. Not toward more powerful assistants, but toward more present ones.
The problem companion robots were trying to solve
Loneliness is not a niche issue. According to one study, nearly one out of three elderly adults resides alone, meaning they do not have daily companions. Research also shows that children whose parents have migrated for work, leaving them in the care of relatives, were 2.5 times more likely to experience loneliness than children whose parents remain with them. Among working adults living alone in urban environments, similar patterns of social isolation emerge, even if they are less visible.
Advertisement
Over the years, technology has time and again attempted to solve this problem via video calls, smart speakers, and messaging apps without much success. Those tools are geared towards communication between people that already have relationships. They do not create presence. They schedule it. That is the gap that a new generation of AI companion robots is being engineered to fill.
Today’s AI robots are different
Today’s companion robots are not just cute and cuddly. They are designed with psychological research, clinical insight and long-term interaction models to be truly useful in real homes.
Three fundamental shifts define the current generation:
From reactive to proactive response. Older robots relied on you speaking to them, but modern robots monitor a room with cameras, microphones, and surroundings sensors to initiate interactions without your input, and they can pick up on your emotions.
From function-oriented to emotion-oriented design. The original pitch for companion robots was about what they could do. The question driving the serious work now is how they make you feel, which is a harder engineering problem and a more honest framing of what the product is actually for.
From standalone hardware to connected ecosystems. Leading brands are creating platforms rather than devices with software included as a built-in layer and remote access from the beginning.
The global AI companion market size was valued at US $36.8 billion in 2025 and is projected to grow from $48 billion in 2026 to $318 billion by 2033, at a compound annual growth rate of 31 percent from 2026 to 2033.
Three household scenarios and interaction models
Ollobot’s advanced AI family companion robot OlloNi SS1 addresses a number of gaps in what existing technology offers.
Advertisement
Elderly individuals living alone. The combination of proactive interaction, fall detection, and persistent presence addresses both safety and companionship without the social overhead of asking family members to check in more frequently.
Children in households where parents work far from home. The SS1 functions as a consistent companion that already knows a child, their preferences, their moods, and their routines. The remote connection features allow parents to stay present without requiring a scheduled call, and the life recording system gives them a passive window into their child’s days that feels less clinical than a monitoring camera.
Single professionals living alone in cities. The SS1 adapts to daily routines, builds up a preference model over time, and provides ambient social presence without demands.
OlloNi SS1 adapts to daily routines over time.Ollobot
What OlloNi SS1 is doing differently?
Ollobot’s goal in building intelligent companion robots is to address the gaps in technology and capability, using innovation not to automate tasks but to fill emotional voids.
Advertisement
Much of the robotics industry has historically pursued human imitation — machines that speak, look, or behave like people. The SS1 is instead designed around familiarity and long-term coexistence rather than realism.
The system integrates multiple subsystems operating in parallel, including visual perception, audio processing, mobility control, and interaction management. It is equipped with a multi-chip AI 4K vision module capable of facial recognition and motion tracking. One small but revealing detail is the inclusion of a physical privacy cover for the camera — a mechanical solution to concerns that software settings alone may not fully resolve.
OlloNi SS1 can actively integrate into family activities, and it can autonomously move closer to capture memorable moments or reposition itself to remain engaged in ongoing interactions.Ollobot
The robot supports advanced mobility across multiple indoor surfaces, including wooden floors, ceramic tiles, and low-pile carpets, with slope climbing capability up to 3.5 degrees. Rather than remaining in a fixed location, it can move naturally throughout the home to stay close to household members as daily activities unfold.
For example, the OlloNi SS1 may greet family members when they arrive home, follow an older adult from the living room to the kitchen while continuing a conversation, remind a child to take a study break after a prolonged period of inactivity, or notice that someone appears unusually quiet and gently check in. During family activities, it can autonomously move closer to capture memorable moments or reposition itself to remain engaged in ongoing interactions.
Advertisement
The robot continues to evolve over time, with over-the-air updates that deliver new features, performance improvements, and AI enhancements
It also incorporates fall detection with optimized accuracy for safety monitoring scenarios. A 6-microphone array enables omnidirectional voice pickup with an effective voice capture range of up to 5 meters, supporting reliable wake-word detection and far-field interaction.
To support continuous companionship, much of the robot’s AI processing takes place directly on the device through its “heart module” architecture, with 16 GB of memory and 64 GB of local storage. This enables the system to retain household memories, recognize familiar faces, and respond with lower latency, making interactions feel more natural even during everyday routines.
Because companion robots are expected to remain available throughout the day rather than only during brief interactions, the SS1 is designed for extended operation, offering up to 12 hours of standby time and around 5 hours of active interaction on a single charge. This allows it to accompany users through meals, conversations, playtime, and other daily activities without frequent interruptions.
Advertisement
To support engaging interactions, much of the robot’s AI processing takes place directly on the device through its “heart module” architecture.Ollobot
Like the relationships it is designed to build, the robot continues to evolve over time. Running on Android OS with over-the-air (OTA) updates, the system continuously receives new features, performance improvements, and AI enhancements, allowing its capabilities to grow alongside the household it serves.
The robot’s behavioral model also improves over time. Rather than reacting to isolated commands, it attempts to establish a baseline understanding of household routines and individuals. Changes in behavior — prolonged quietness, unusual inactivity, or emotional cues — become triggers for interaction.
Presence instead of utility
Several features in the OlloNi SS1 illustrate this emphasis on presence and continuity in its interactions.
The system can identify different household members, including pets, and adapt responses accordingly. Remote communication features allow family members to connect through the device without treating every interaction like a scheduled call. Environmental sensors support contextual reminders tied to weather or room conditions.
Advertisement
Its “2+1” multi-display configuration is also designed around emotional communication. Two circular side displays function as expressive “emotional eyes,” while a separate primary display handles information and structured interaction. The separation allows emotional signaling and functional communication to operate independently, creating more intuitive nonverbal interaction even when no dialogue is taking place.
The robot’s behavioral model improves over time. Rather than reacting to isolated commands, it attempts to establish a baseline understanding of household routines and individuals.Ollobot
The SS1 also includes an automated life-recording system built on facial recognition and behavioral-event detection that can capture moments such as laughter, physical closeness, or group interaction automatically. An integrated AI vlog engine can then organize those moments into edited short-form videos with automated sequencing and soundtrack generation. The design intent is to preserve spontaneous domestic moments without requiring active documentation behavior from users.
An integrated AI vlog engine can organize recorded moments into edited short-form videos with automated sequencing and soundtrack generation
Visual data is processed primarily on the device through the SS1’s on-device AI architecture, with household memories stored locally and managed within Ollobot’s proprietary ecosystem instead of being shared with third-party smart home platforms. Access to recordings and live feeds is restricted to authorized users through the companion app, while encrypted communication helps protect data during remote access. Users also retain direct control over recording preferences, and the physical camera privacy cover provides an additional hardware-level safeguard whenever visual monitoring is not desired.
Advertisement
Remote communication is similarly structured around persistence rather than transaction. Traditional video calls are episodic and screen-bound; the SS1 instead acts as a continuously present interface embedded inside the household environment. Through autonomous mobility, environmental awareness, and persistent household memory, remote family members interact with an ongoing domestic context.
The larger shift to “gentle intelligence”
Ultimately, gentle intelligence is not about making robots behave more like humans — it is about helping them fit more naturally into human lives. Each OlloNi SS1 unit develops a unique behavioral profile based on its household. Two units running in different homes for a year will have become meaningfully different from each other, shaped by the specific people, habits, and rhythms of where they live.
That kind of long-term personalization is what early companion robots never had. It is also what makes the difference between a product that ends up on a shelf and one that actually earns its place in a home.
Bill Gates at the keyboard in a 2018 file photo. (Gates Notes Photo)
Bill Gates is legendary, bordering on notorious, for his late-night emails — missives to colleagues with piercing questions about Java back in the day, or malaria these days, or whatever esoteric topic he happens to seize upon at any given moment.
But increasingly, he is sending these messages to AI, not to people. He’ll bounce something off Claude, get ChatGPT to weigh in, and insert himself in the middle.
He described the pattern in an interview with GeekWire: “It’s 3 a.m., I want to understand sodium batteries. Now, there’s no reason to go to sleep. Here we go! Yeah, it’s crazy.”
If you’re a curious person, he said, “this is a mind-blowing time.”
In terms of productivity, he added, “we are in heaven.”
Advertisement
All of which might be predictable. This is Bill Gates, after all. Now 70 years old, he has spent more than five decades impatient for the future to arrive — making the case that innovation, on the whole, will ultimately put humanity and the world in a better place.
So here’s the surprise twist: He’s now deeply concerned about where technology is headed, how fast it’s progressing, and how little the world is doing to get ready.
In a new essay, Gates says the “turbulent AI era” has arrived, with technology threatening to erase categories of jobs, supercharge fraud and deepfakes, lower the bar for cyberattacks on critical infrastructure, make it easier to engineer a deadly new disease, let governments kill without humans involved in the decision, and fundamentally change how kids grow up.
If someone came up with a credible plan to slow the pace of AI globally, he writes, he’d likely support it. But he doesn’t expect one. The geopolitical and economic forces are too much.
Advertisement
He says that the world needs to take action, and offers three ideas to start:
Build new institutions, at home and globally. No existing agency was designed for a technology that touches jobs, security, health, energy and elections all at once, he writes.
Gates calls for new national bodies that can set priorities across agencies, plus a new international organization modeled on nuclear weapons inspections, aviation rules and the ozone treaties.
Set aside jobs for humans. Gates calls this “Human Reserved”: work that machines will be fully capable of doing, but that we decide to keep for people anyway. The model is a nature reserve — land where we could build roads and buildings, but choose not to, because the loss would be too great.
Advertisement
One example: a robot delivering the news that you have an incurable disease. “There’s no technical reason why it couldn’t,” he writes. “Yet it shouldn’t.”
The idea came in part from watching the caregivers who looked after his father through Alzheimer’s, work he describes as “irreplaceably human.”
Tax AI tokens and robots. Today a company that hires a worker pays payroll taxes, while a company that buys a robot deducts the cost. Gates says that gives employers a reason to replace people. He’s calling for a tax on AI to change the incentives and help pay for retraining.
He first floated a robot tax nine years ago, but the idea was widely dismissed. He’s still for it. He acknowledges that it isn’t economically efficient, but says that with innovation accelerating, we can afford a little inefficiency as the price of keeping people employed.
Advertisement
Gates is candid that he doesn’t have all the answers, particularly on the proposal for “Human Reserved” jobs. Who decides what gets reserved, and by what criteria? How do you keep companies from using robots in the jobs that are supposed to stay human?
These, he writes, “will need to be worked out in public.”
In the meantime, he’s working it out with Claude. Gates said he has talked the idea through with the chatbot, thinking through different ways to get the share of work reserved for humans up to 40%, using shorter workdays and earlier retirement to spread what’s left around.
Crossing the threshold
In the GeekWire interview, Gates said the essay came out of a specific realization: the AI industry is blowing past its own warning signs, one after another, and almost nobody is saying so out loud.
Advertisement
For years, he said, people in AI described certain moments as dangerous points where the industry would stop and think hard before going further: making it easier to build a bioweapon, making it easier to launch a cyberattack, building machines people become emotionally dependent on, wiping out large numbers of jobs, and losing control of the technology itself.
“We’re in the process of crossing every single one of those thresholds,” he said.
Meanwhile, nobody in the industry wants to be first to step on the brakes. “Most people you talk to will say, yeah, well, if everybody else would slow down, maybe I would, too,” he said.
Gates said one way out of that standoff is for governments to step in.
Advertisement
His example: any AI model capable of designing new molecules — the capability that would let someone engineer a new disease — should be monitored. The monitoring would be mandatory rather than voluntary, and it would cover free models as well as commercial ones. It would also have to be written so a company can’t copy the model elsewhere and strip the monitoring out.
“To me, that’s kind of like common sense,” he said. “But we don’t see a specific proposal to do that.”
‘The whole thing seems so empty to me’
Under an executive order signed by President Trump in June, AI companies are asked to submit their most powerful models for government testing up to 30 days before release. The order specifically bars the program from becoming a licensing or preclearance requirement. The White House finalized the framework in early August.
Gates said he doesn’t get it.
Advertisement
“What is the threshold that’s being examined, and what is the action taken when you cross that threshold?” he said. “The whole thing seems so empty to me.”
If the world can’t take these basic steps, he said, “I really am going to throw up my hands.”
If the process stays voluntary, with no line and no consequence for crossing it, “we’re going to look back on this as a kind of eye-of-the-storm type moment,” he said.
Asked if he had taken his proposals to the Trump administration or to other heads of state, Gates said with a bemused tone, “Well, you could tell me who at the White House I should be talking to about this.” He said he hopes the essay reaches people in Congress and in the executive branch.
Advertisement
He said the public argument among AI companies over whether the risks are real is beside the point, because privately the people running them already agree. “I know they’re all worried,” he said, “or all of them that I know, which is basically everybody but Elon.”
People inside AI companies who acknowledge the downsides, Gates said, get told: “Hey, you’re hurting our PR while we’re trying to raise trillions of dollars.”
Gates said he previously expected losing control of AI to be a distant problem, something to worry about “many years from now.” He’s no longer convinced that’s the case.
He referenced an Aug. 11 episode of the Dwarkesh Patel podcast featuring Ryan Greenblatt, chief scientist at the AI safety group Redwood Research. Greenblatt said that as AI systems get more capable, the people building them understand less and less about what is happening inside, and that sufficiently advanced models could end up working against their creators.
“These are people who are super expert on the thing, going, well, maybe we won’t be able to control these things,” Gates said. “I mean, what kind of risk have we chosen to run here?”
Advertisement
In the poorest countries, he expects AI to do more good than harm. In the countries where the Gates Foundation works, doctors, teachers and farm advisors are all in short supply. AI can help fill those gaps. The foundation will lay out that work at its Goalkeepers event next month, including an effort to make AI models work as well in African languages as they do in English.
The job losses, he added, will hit rich countries first.
It’s the first big wave of new attention on the Microsoft co-founder and Gates Foundation chair since he answered lawmakers’ questions in the Jeffrey Epstein investigation on June 10, sitting for a nearly six-hour voluntary interview with the House Oversight Committee.
Advertisement
Gates, who has not been accused of any wrongdoing, was asked by Axios whether he’s concerned that the Epstein issue could undercut his message. According to the site, he compared this to earlier situations when personal and professional challenges diminished his ability to speak out on key subjects: during the Microsoft antitrust trial, and his divorce from Melinda French Gates.
The AI Road Ahead
For all of this, Gates is still thinking about how technology will change human life and productivity, in many ways for the better on an individual level.
A key step, he said, will be establishing broad-based persistent memory for AI agents across contexts. For now, AI still doesn’t know you like a human assistant who’s familiar with your relationships and how you think about your time.
Gates sees the role of apps changing in the future. Instead of bouncing between different pieces of software, he said, AI will increasingly be the primary interface. “You won’t go to those applications,” he said. “You’ll just go to your personal agent.”
He also sees AI continuing to transform shopping, to an extreme: “We will get to a point where you won’t buy things yourself. You just won’t.” Telling the agent to help you buy something, “it’ll consider so many more things, and it’ll make it so much easier for you to do it.”
Asked whether he is still an optimist, Gates didn’t answer directly. “I don’t think being pessimistic is helpful,” he said.
“I do think, wow, this is sure an interesting time. I’m the guy who in my 30s thought people in their 50s or 60s didn’t understand anything.” He called it “kind of bizarre” that he would be delivering a message like this at 70.
“But I am very concerned. And honestly, when you get people one-on-one, so are they.”
If you’re the type of person who’s drawn to the idea of open-ear audio as well as a standout design, then there’s plenty here to catch your eye. The brushed metallic finish and looping clip-around design make these feel more like a fashion accessory than earbuds, and with several colour options to choose from, there’s room to customise them to your look.
Each bud attaches to your earlobe via a silicone-coated flex arm. Because they don’t seal your ear canal at all, you can wear them for hours without any of the fatigue you’d get from a more traditional pair of in-ears. Fitting them is straightforward, and nudging them up or down to find your sweet spot takes no effort at all. You also get IPX4 water resistance for sweat and splashes, which makes them a reasonable companion for outdoor activities.
On the features front, the Bose Ultra Open Earbuds run on Bluetooth 5.3 with SBC, AAC and aptX Adaptive codec support. Snapdragon Sound is along for the ride too, and a later update brought Bluetooth multipoint, so you can be paired to two devices at the same time.
Wireless performance held up well during testing, with barely any dropouts to speak of. Bose’s Immersive Audio mode is worth exploring as well but only if you’re in a quiet environment. You can pick between Still and Motion modes depending on whether you’d rather the sound stay fixed or shift with your head movements, though turning it on does eat into battery life.
Advertisement
That battery life is rated at 7.5 hours per charge, beating the original Sony LinkBuds and is decently respectable for day-to-day use. Flipping on Immersive Audio drops that figure to around 4.5 hours. Fast charging means a two-minute top-up is good for two hours of playback. Wireless charging is available, but it requires a separate slip-on accessory, which feels like an odd omission at £299.
Sound quality, as ever with open-ear designs, comes with caveats. In the right setting, the Bose Ultra Open Earbuds sound clear and clean in the mids and highs. Bass, predictably, is where the open design works against them, and in louder environments like a packed commute or a busy gym, the volume ceiling becomes noticeable.
Immersive Audio widens the soundstage but softens the overall clarity, so sticking with plain stereo tends to deliver a more satisfying result. Call quality is also a disappointment, as background noise has a habit of creeping back in the moment you start speaking, which muddies your voice for whoever’s on the other end.
If you’re already set on open-ear audio and you want something that’s comfortable, distinctive and capable, the Bose Ultra Open Earbuds are well worth considering. Just go in knowing that the bass is thin and call quality isn’t their strong suit, and you won’t be left feeling short-changed.
Speed over Form: Qualcomm is getting ready to spill the beans about the next generation of its computing products. The fabless chip designer has introduced a new, powerful CPU core architecture that’s allegedly reaching unprecedented frequency rates and IPC performance for a power-constrained mobile processor.
Qualcomm is pushing its Oryon CPU up to 5GHz, which should provide a significant performance uplift on the next generation of mobile flagship devices. The company said the upcoming Oryon architecture will be the world’s fastest mobile CPU, although it will require some additional silicon magic to truly shine in AI workloads and other trendy computing tasks.
Qualcomm first introduced the Oryon CPU core in 2024, together with the first generation of the Snapdragon X SoC series. The core implements the Arm architecture within a custom Qualcomm design, and the next-gen Oryon CPU will apparently push this architecture customization effort even further.
Aside from the improvements coming from a newer process node, upcoming Oryon cores will implement a fully custom chip design. Qualcomm attributes the 5GHz milestone to its in-house Arm microarchitecture, the implementation choices behind the core, and the subsystem managing everything else around the main computing cores.
Advertisement
According to the single slide provided by Qualcomm, the next-gen Snapdragon SoC will include two Oryon “Prime Core” units running at 5GHz alongside six other “Performance Core” units.
Furthermore, the Prime cores will boast strong instructions-per-clock (IPC) performance to actually accelerate real-world usage scenarios. Needless to say, Qualcomm is mostly thinking about agentic AI tasks being orchestrated by the on-device CPU.
Another crucial element in Qualcomm’s next-gen SoC is a new cache architecture named FlexCache. The new technology gives “heterogeneous” cores access to cached data (or rather, the same cache pool), performing dynamic cache allocation based on the most pressing workload at hand.
The 5GHz Prime Cores will be able to draw on the entire cache pool when a computing task demands it, Qualcomm explains. This way, larger tasks can stay “resident” within the processor’s cache memory without resorting to the slower system RAM outside the SoC.
Advertisement
In a world where memory chips have become the most sought-after silicon products, Qualcomm hopes that FlexCache will significantly improve performance even on devices with limited (system) RAM capacity.
Qualcomm said that FlexCache will give the 5GHz Oryon CPU the performance edge over competing SoC cores. The company highlights some specific scenarios for this alleged speed uplift, which include agentic AI, gaming, app multitasking, and video editing.
Something to look forward to: CD Projekt Red had already promised plenty of Witcher news at Gamescom, and it delivered more than expected. The studio has announced The Witcher 3: Wild Hunt Remastered, a substantial overhaul launching September 29 as a free upgrade for existing owners. It also provided the first proper look at Songs of the Past, a new paid expansion arriving in 2027.
The Witcher 3 Remastered appears to go considerably further than simply sharpening up the visuals of the 2015 RPG. CD Projekt Red says the new edition will update graphics, performance, combat, movement, progression, and several other parts of the game.
Visual improvements include upgraded textures for environments, characters, and monsters, along with denser foliage and revised lighting, rendering, and ray tracing. Combat is also being reworked with new animations, finishing moves, and a more adaptive camera.
Other changes include improved traversal, revised horse controls, enhanced monster behavior, new Witcher Sign effects, a redesigned skill tree, transmogrification, an expanded photo mode, and cross-platform mod support.
The upgrade launches September 29 on PC, PlayStation 5, and Xbox Series. The Witcher 3 will also receive a native Nintendo Switch 2 version, while CD Projekt Red is bringing the game to Battle.net through a new partnership with Blizzard.
Existing owners on eligible platforms will receive Remastered at no additional cost.
CD Projekt Red had another surprise for anyone who only owns the base game. Hearts of Stone and Blood and Wine, the two acclaimed expansions released in 2015 and 2016, are now being made available to existing Witcher 3 owners for free. That means players can revisit essentially the complete current Witcher 3 experience before another major adventure arrives next year.
“Songs of the Past” revealed
CD Projekt Red also showed the first footage from Songs of the Past, The Witcher 3’s first new expansion since Blood and Wine a decade ago. The expansion sends Geralt to Letten, a previously unseen region portrayed as a peaceful rural land built around farming, festivals, and local folklore. Naturally, things aren’t quite as idyllic as they initially appear, with an old threat lurking beneath the surface.
Geralt will encounter new enemies, reconnect with familiar characters, and gain additional abilities. One of the more noticeable additions shown so far is a new chain weapon, which should give Geralt another option for controlling and attacking enemies.
Songs of the Past is being developed by CD Projekt Red alongside Fool’s Theory, the Polish studio behind The Thaumaturge. Fool’s Theory is also leading development of the remake of the original Witcher. Songs of the Past will be a paid expansion. It will launch sometime in 2027 for PC, PlayStation 5, Xbox Series, and Nintendo Switch 2.
CD Projekt Red previously indicated that the expansion would be comparable in scope to Blood and Wine, which introduced the large Toussaint region and added dozens of hours of new content.
The company plans to reveal considerably more during Gamescom. Five days of Witcher-focused REDstreams are scheduled from August 26 through August 30, including gameplay presentations, interviews, and behind-the-scenes material covering both Remastered and Songs of the Past.
Advertisement
All of this will help fill what remains a fairly long wait for the franchise’s next numbered installment.
The Witcher 4, which shifts the spotlight from Geralt to Ciri, is currently targeting a 2028 release. CD Projekt Red co-CEO Michał Nowakowski emphasized that 2028 remains a target rather than a firm launch date, leaving room for the project to move if additional development time is required.
Whenever The Witcher 4 does arrive, it is expected to begin an unusually busy period for the studio. CD Projekt Red has said it wants to release three Witcher games within six years, with its collaboration with Epic Games on Unreal Engine 5 intended to help shorten development cycles between installments.
Beyond The Witcher 4 and the original game’s remake, the company has another Witcher project in development, alongside the next Cyberpunk game and an entirely new IP.
Advertisement
For now, though, CD Projekt Red is making its 11-year-old RPG the center of attention again. Between a free overhaul in September and a full expansion next year, The Witcher 3 is getting a much bigger second life than anyone would have expected this far after launch.
Findings by Consumer Reports and JD Power suggest that plug-in hybrids are far less reliable than their internal combustion counterparts, but digging into the numbers and research methodology reveals it’s not quite that simple. First, although Consumer Reports’ survey found owners of electric vehicles and plug-in electric vehicles report about 80% more problems on average than owners of internal combustion engine vehicles, it lumped EVs and PHEVs together. Without breaking out the PHEV number, it’s hardly a definitive answer, and it’s further complicated by there being far fewer EV and PHEV models on the market than ICE vehicles, which means a few bad apples can seriously skew the data.
When it comes to assessing reliability, it’s also crucial to consider how the data is sourced. Consumer Reports surveyed owners of about 380,000 vehicles from its own subscriber base, which means most of its data was self-reported. That’s a problem, because it means it’s highly anecdotal. As with online reviews, disgruntled owners tend to be more inclined to respond to surveys than satisfied ones. JD Power’s 2026 reliability report has similar problems, in that it also relies heavily on self-reported information, but it recruits respondents using vehicle registration and ownership information, rather than an existing subscriber base. Additionally, JD Power began collecting repair data from dealerships in April 2026, which will add credibility to subsequent reports.
Advertisement
Some PHEVs are less reliable than others
Gerenme/Getty Images
However, regardless of these methodology caveats, both Consumer Reports and JD Power found a significant difference between the reliability (or perceived reliability) of PHEVs and other vehicles. But further digging suggests how well a PHEV performs varies enormously between makes and models, and can also vary substantially by powertrain within the same model family.
For instance, the Mazda CX-90 is available as both a gas-powered inline-six and as a PHEV. We said of the PHEV variant that the mixture of power sources can leave, “the two halves of the drivetrain occasionally feeling like they’re fighting rather than collaborating.” Meanwhile, we scored the ICE variant 9/10 and said, “it’s Mazda’s overall package that leaves its sizable SUV feeling genuinely special.” But don’t just take our word for it, Edmunds was similarly impressed by the gas-powered model, but underwhelmed by its PHEV sibling.
Conversely, Toyota’s Rav4 Prime PHEV consistently bests its ICE stablemate in reviews, with the 2021 model year iteration managing 87/100 on JD Power’s quality and reliability score (and an overall score of 82/100), outperforming other compact SUVs like the Jeep Cherokee and Honda CR-V in the process. Moreover, the SlashGear team has consistently found Toyota’s PHEVs highly reliable, and the RAV4 Prime even made our list of nine of the most reliable SUV models made by Toyota. PHEVs as a category may be less reliable than other powertrain options, but within their ranks, which one you choose makes a huge difference.
Nvidia is expected to report quarterly revenue of $92.18bn, close to double the same period last year and its fastest growth in seven quarters, with analysts forecasting a further rise to $104.20bn in the current quarter.
The numbers are not really the issue anymore, because the scrutiny has moved to how much of that demand the company is underwriting itself after arranging $500bn of financing for customers this month.
The commercial question for the quarter is the transition to Rubin. Shipments of the next-generation processors begin this autumn, and how quickly customers move off Blackwell determines whether the guidance holds.
Morgan Stanley expects Rubin to contribute nearly $9bn in the current quarter, on the argument that it unlocks a large improvement in what the bank calls AI factory economics. Nvidia has already said Vera Rubin is in full production with OpenAI deploying at scale.
Advertisement
The financing arrangements are where the analyst questions have concentrated. Alongside the $500bn package, Nvidia guaranteed up to $105bn for OpenAI’s Ohio data centre, which places the chipmaker somewhere between supplier and lender to its own customers.
“This makes them a kind of central banking figure in the AI space,” said Brian Mulberry of Zacks Investment Management. “The real risk is total AI exposure with no diversification.”
Credit markets have already registered the point. Nvidia’s roughly $750bn of announced AI commitments pushed its own credit default swaps to record levels, which is the bond market saying it would like to see the demand stand on its own.
The structural worry is not that customers cannot pay. It is that a supplier financing purchases of its own product records revenue that looks independent of it, and nobody outside the company can easily separate the two from the income statement.
Advertisement
Customer concentration compounds it. A handful of hyperscalers and model labs account for a very large share of Nvidia’s revenue, and several of them appear on both sides of the ledger as buyers and as recipients of its investment or financing.
Competition is the other pressure, and it is no longer theoretical. Amazon, Google, and Meta all have custom silicon programmes; AMD and Intel are pushing on inference, and inference is the workload that grows fastest once models are deployed rather than trained.
Inference also has different economics to training. It rewards cost per token rather than raw throughput, which is the ground on which custom silicon competes best and where Nvidia’s architectural advantage is least decisive.
Market sentiment has been unusually mixed for a company posting these figures. Nvidia shares are up 11.8% for the year but have lagged the major indices recently, and it briefly lost its position as the world’s most valuable company to Apple.
Advertisement
Cost pressure is now visible in Nvidia’s own pricing too. The company has told customers that AI server prices are rising by more than 15% because of memory shortages, which passes the DRAM crunch straight to the buyers it is simultaneously helping to finance.
The bull case has not gone anywhere in the meantime. Demand for inference capacity keeps outrunning supply; the buildout has years of announced projects behind it, and no competitor has yet matched the software stack that keeps customers on Nvidia hardware once they are there.
The guidance will carry more weight than the reported quarter. A beat on revenue with a soft outlook would confirm what the credit market has been signalling, and a strong outlook invites the follow-up question of how much of it is committed capacity the company arranged the money for.
Neither answer settles the underlying argument. Nvidia can keep growing at this rate for as long as the buildout continues, and the buildout continues for as long as capital keeps arriving, which is a circular sentence describing a circular arrangement.
In an uncharacteristic August launch, Apple has updated the Mac mini, retaining the same design, adding the M6 and M5 Pro processors, and a new record-high starting price.
The release comes almost two years after the then-redesigned M4 Mac mini was released. There’s nothing different externally, with the major change being the M6 processor.
And it comes after wait times were as long as two months on the morning of August 25 for the M4 version of the hardware. It will almost certainly sell out rapidly upon release, and we advise that if you want one, you order one when the preorder period starts.
“Today, we’re debuting the next giant leap in performance and AI compute for Apple silicon with the incredibly advanced M6 and the most powerful M-series chip yet, M5 Ultra,” said Sri Santhanam, Apple’s vice president of Silicon Engineering Group. “Built using the cutting-edge 2 nm process, M6 combines a new CPU complex, two additional CPU and GPU cores, a Dual 16-core Neural Engine, and more unified memory bandwidth to power through workloads with amazing energy efficiency.”
Advertisement
“And for the ultimate desktop performance and the ability to run massive AI models, M5 Ultra features a massive GPU, now with Neural Accelerators, and more unified memory bandwidth, pushing the boundaries of what a desktop can do.”
M6 has a 12-core CPU with two super cores, four performance cores, and six efficiency cores. Apple says the combo delivers up to 1.2x faster multithreaded performance as compared to M5, and up to 2.4x faster than M1.
Also, the M6 features a 12-core GPU, each with an accompanying Neural Accelerator. Apple also says that this delivers a nearly 30 percent increase in peak GPU compute for AI compared to M5, and more than eight times faster compared to M1.
“Mac mini has always been our most versatile Mac. Whether it’s being used as a home computer, powering a professional studio, or as an always-on agentic device, it’s the little Mac that can do it all,” said Johny Srouji, Apple’s chief hardware officer.
“Today, we’re taking this versatility even further. With a more powerful CPU and graphics, Neural Accelerators in the GPU, and higher memory bandwidth, Mac mini with M6 delivers a whole new level of AI performance. And with up to an 18-core CPU and 20-core GPU, Mac mini with M5 Pro is an ultracompact powerhouse for complex pro workflows. We can’t wait to see all the incredible ways people will use it next.”
Advertisement
The new M6 chip supports up to 32GB of unified memory. That RAM also provides up to 170GB/s of unified memory bandwidth.
Both Mac mini M5 and Mac mini M5 Pro models include Wi-Fi 7 and Bluetooth 6, as well as upgraded 2.5Gb Ethernet. A 10Gb option is available at an extra cost.
Thunderbolt 5 is not universal. On the back, there are three Thunderbolt 4 ports on the Mac mini with M6. The Mac mini with M5 Pro has three Thunderbolt 5 ports, along with HDMI and Ethernet.
At the moment, there is no sign of the first units being made in America. At some point later in 2026 they are expected to be assembled in Houston, TX.
Advertisement
Pricing and availability
Customers can preorder the new Mac mini with M6 and M5 Pro starting August 25. Neither model ships until September 22.
Mac mini with M6, 12-core CPU, 12-core GPU, dual 16-core Neural Engine, 16GB of RAM, and a 256GB SSD starts at $899.
Mac mini with M5 Pro, 15-core CPU, 16-core GPU, 24GB of RAM, and a 512GB SSD starts at $1,699.
A $100 discount is available on both models for US military members and in the education store. The $899 price is the Mac mini’s highest starting price yet.
‘Customers increasingly … [use] Mac mini as a powerful platform for agentic AI’, Tim Cook said in July.
After rumours of an impending announcement were published this morning, Apple has officially revealed its first new Mac mini desktop computer in nearly two years, with two new models featuring the company’s new M6 and M5 Pro chips respectively.
With the M6 chip – also newly announced today (25 August) – the Mac mini can deliver up to four-times faster AI performance, two-times faster storage and graphics, and 40pc faster CPU performance, according to Apple.
The M6 is the company’s first 2nm chip, which features a larger 12-core CPU complex with “the world’s fastest CPU core” according to Apple, along with a larger 12-core GPU with neural accelerators, a dual 16-core neural engine, and up to 170 gigabits per second (Gb/s) of unified memory bandwidth.
Advertisement
Meanwhile, the M5 Pro features up to an 18-core CPU and up to a 20-core GPU – with neural accelerators in each GPU core – with an “enhanced shader core and third-generation ray tracing for complex work” in 3D design, VFX, and game development. The Mac mini with the M5 Pro also supports up to 64GB of unified memory and 307Gb/s of memory bandwidth.
The Mac mini with M6 is priced from $899, while the version with the M5 Pro is priced at $1,699.
The new Mac mini was announced after Bloomberg reported on its impending reveal earlier today.
Sources told the publication earlier today that the new device was expected in the coming days – the tech giant has since officially unveiled the device.
The upgrade to Apple’s Mac mini line follows a growth in demand for the small, portable device popularly used to run AI applications locally.
“We’re seeing customers increasingly … [use] Mac mini as a powerful platform for agentic AI [or] deploying clusters of Mac Studio systems to run frontier-class models locally,” Cook said at the company’s last earnings call in July.
The machine is available for pre-order starting today, according to the company, with the product set to launch on 22 September.
Advertisement
The Mac mini is one of a number of Apple products set to launch soon, including the iPhone 18 Pro, Apple’s first foldable smartphone, a new range of Apple Watches, updated AirPods and a smart home device.
The company is also readying a new Apple TV set-top box, a HomePod mini smart speaker and a revamped iPad mini with an OLED screen.
“The rapid expansion of AI data centres has created an extraordinary surge in demand for memory and storage”, a company spokesperson said at the time, adding that Apple had “never seen a component price increase this much, this quickly”.
Advertisement
According to Counterpoint Research, global smartphone shipments are poised to drop nearly 14pc in 2026, with prices expected to jump by as much as 13pc.
The Mac mini has been a particular problem for Apple in supply chain terms, with deliveries taking weeks or months to be completed. Cook said that the company’s new advanced manufacturing centre in Houston, Texas will be assembling Mac minis starting “later this year”.
With additional reporting by Colin Ryan
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
Advertisement
Updated, 3.55pm, 25 August 2026: This article was amended to include up-to-date information on the new Mac mini, as the device was officially announced shortly after our article was published.
Apple’s top-tier Mac, the Mac Studio, has been given a big AI bump with the M5 Ultra. Here’s how it compares against the M3 Ultra model it replaces.
The Mac Studio now has M5 Ultra, an upgrade from the M3 Ultra.
The new Mac Studio arrived on Tuesday, with the all-new M5 Ultra now the top-of-the-range chip for tough productivity workloads. Apple is positioning it squarely as a desktop for doing a lot of AI-related processing, including running models locally. The design stays faithful to the outgoing model, so the story lives inside the chip. This time, the headline is artificial intelligence rather than raw core counts. Continue Reading on AppleInsider | Discuss on our Forums
SpaceX plans to spend up to $100 billion building a new Starship launch facility in Vermilion Parish, Louisiana, which the state says could eventually “support thousands of launches annually.” CNBC reports: Once built, Starbase, LA will serve as the main launch facility for Starship, SpaceX’s in-development rocket that’s the largest ever built and designed to be fully reusable. Musk has aspirations to massively expand SpaceX’s Starlink satellite network, to launch orbital data centers and to someday colonize Mars. Those ambitions, and justification for SpaceX’s nearly $2 trillion market cap, hinge on the success of Starship.
SpaceX said, in a video posted to its website on Tuesday, that the search for a site to build launch pads and have access to ample fuel for its operations took about seven years. Musk said in the video that the company would bring “probably 10,000 really exciting jobs” to Louisiana with the spaceport. A fact sheet posted online by the state of Louisiana said SpaceX is expected to generate an estimated “3,000 direct new jobs over the next 10 years” in the state, and “8,100 indirect new jobs.” SpaceX said it would work on a “historic coastal restoration” in Louisiana, in partnership with the state to “bring the marsh back” and ensure the wetlands around its new facility will have “storm protection.”
OlloNi SS1 adapts to daily routines over time.Ollobot
OlloNi SS1 can actively integrate into family activities, and it can autonomously move closer to capture memorable moments or reposition itself to remain engaged in ongoing interactions.Ollobot
To support engaging interactions, much of the robot’s AI processing takes place directly on the device through its “heart module” architecture.Ollobot
The robot’s behavioral model improves over time. Rather than reacting to isolated commands, it attempts to establish a baseline understanding of household routines and individuals.Ollobot
You must be logged in to post a comment Login