Always AI’s Fault: IBM took the unusual step of warning shareholders ahead of its financial results for the second quarter of 2026. Revenue is still growing slightly, but the industry’s AI-driven FOMO is expected to significantly affect the company’s prospects across some of its most important product lines.

In a recent letter to investors, IBM CEO Arvind Krishna explained that the company experienced some unexpected shortfalls in its latest financial performance. The former pioneer of the personal computer industry now primarily sells software, cloud services, and mainframe systems, but potential customers have recently scaled back major mainframe investments because the AI boom is pushing companies to prioritize spending on memory chips, storage drives, and GPUs instead.

According to the preliminary figures shared in Krishna’s letter, IBM remains in solid shape. The company’s revenue is expected to grow by 1% in the second quarter of 2026, reaching $17.2 billion. Software revenue is projected to increase by 5%, while Infrastructure revenue is expected to decline by 7%.

IBM recently prepared the launch of its new Z17 mainframe, and the company had already expected Infrastructure revenue to decline by a “low-singlet” digit percentage in the second quarter and throughout the year. However, according to Krishna, clients decided in the final weeks of June to shift corporate spending away from IBM mainframes and toward servers, storage, and memory products.

Customers were apparently pushed to secure additional hardware components before Big Tech companies and AI firms could acquire every memory chip and hard disk drive available. As SoftBank CEO Masayoshi Son recently said, the AI industry is expected to continue investing trillions of dollars in new and unprecedented data center projects, despite growing signs that the AI bubble could eventually begin to unravel.

This AI-driven shift had a significant impact on IBM’s client spending patterns, Krishna said. The US company had already anticipated some supply chain-related disruptions but did not expect the scale of the change in customers’ capital expenditure priorities. Other factors affecting Big Blue’s mainframe (Infrastructure) business include cybersecurity concerns and the company’s inability to quickly adapt to rapidly changing market conditions.

Following Krishna’s letter, IBM’s stock suffered a 25% decline, marking the company’s worst single-day performance in history. Preliminary revenue figures came in below analysts’ expectations, with analysts forecasting that the quarter would close at $17.86 billion.

At any rate, IBM is still highlighting some positive aspects of its financial results despite the overall disappointing quarter.

Big Blue’s strongest-performing products include Red Hat, which is expected to grow by 11% compared with the previous quarter. The company is also launching Lightwell, a service designed to leverage Anthropic’s Mythos AI frontier technology to improve open-source security. Customers will need to pay a subscription fee to access it.

Apple sued OpenAI last week over alleged hardware trade secrets theft.

OpenAI is pitting its AI capabilities against the likes of Amazon and Google with its very first consumer gadget, a “mobile, screen-free smart speaker”, according to a Bloomberg report.

The publication’s sources said the new device is designed to be a next-generation, ChatGPT-powered home computer, and meant to be a physical embodiment of the AI chatbot.

Sources told journalist Mark Gurman that the device is meant to be a “human-like AI companion” that stays at home and controls smart-home appliances, responds to queries and plays media – tapping into OpenAI’s chart-leading ChatGPT models for these capabilities.

Similar devices, such as Amazon Alexa, Apple Home and Google Home, are already widely available on the market, but OpenAI is seemingly attempting to make upgrades to existing model standards. According to Bloomberg’s report, OpenAI is hinging on its ability to connect with users on a “human-like level”.

The new smart speaker is expected to become “increasingly personalised” over time as it continues to gather data on its users, including via reading emails.

“The speaker incorporates mechanical elements that can move on their own, creating a sense that it is alive and not just an object responding to commands,” read the report.

Though described as a speaker, the device, which is still under development, will include a camera along with other sensors to help it better understand context, sources said. It will also come with a rechargeable battery.

The speaker will communicate via the new ChatGPT voice model GPT-Live that was rolled out earlier this month. According to the AI giant, the model is capable of speaking with a cadence similar to humans, and can listen and speak at the same time.

GPT-Live “make[s] talking with AI feel much more like having a real conversation”, OpenAI said at the time of its launch.

Unlike its biggest direct competitor, Anthropic, which has a narrowed focus on the enterprise AI market, OpenAI is expanding its scope into the heavily populated consumer gadgets market.

The tech giant, which has shaky plans to go public this year, has shut down less profitable ventures such as the Sora video generation model and its standalone AI browser Atlas.

The company spent $6.5bn last summer to acquire Io, a hardware start-up co-founded by Apple design veteran Jony Ive and OpenAI’s current chief hardware officer Tang Yew Tan. And by November, it had finished developing its first device prototypes.

Its hardware division is currently developing five different products, but hopes to one day create a mobile AI device capable of replacing the smartphone.

The news comes as Apple, in a fresh lawsuit last week, accused OpenAI and its hardware division of stealing trade secrets.

The iPhone maker named Tan, its former VP of product design for iPhone and Apple Watches, in the suit, and accused OpenAI of taking part in a “coordinated pattern of misconduct” to gain access to confidential information about its unreleased products.

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.

Google’s new Home Speaker was already more interesting than the average $99 smart speaker because it was the company’s first audio product designed specifically around Gemini for Home. It can now do something far more useful than answer questions, control the lights, or quietly confirm that Orwell lacked imagination about consumer electronics.

One or two Google Home Speakers can connect wirelessly to the Google TV Streamer and become its primary audio system. A single speaker replaces the television’s built-in sound, while two speakers can operate as a stereo pair and enable Google’s new Spatial Audio processing.

The complete two-speaker setup costs $299.97 before tax: $99.99 for the Google TV Streamer and $99.99 for each Home Speaker. Existing Google TV Streamer owners can add the stereo pair for $199.98. That puts the system below most premium soundbars and makes it potentially useful for bedrooms, apartments, offices, and smaller living rooms where an AVR, subwoofer, and several loudspeakers would be excessive.

Related Reviews:

Google Finally Connects Its Speakers to the Television

Google has had smart speakers, streaming hardware, multiroom audio, and voice control for years, but it never combined them into a particularly convincing television sound system.

The 2026 Google Home Speaker begins to close that gap. Each compact speaker uses a single 58mm full-range driver designed for omnidirectional playback. It also supports Wi-Fi 6, Bluetooth 5.4, Thread 1.3, Matter, and Google Cast speaker groups, while Gemini for Home handles voice commands and smart-home control.

Setup is handled through either the Google TV interface or Google Home app. Users can select up to two speakers, place them beside the television, assign the left channel, and activate Spatial Audio. The setup process asks for the approximate viewing distance and the distance between the speakers so Google can adjust its processing to the room.

Google says the two-speaker system simulates surround sound and creates the sensation that audio is arriving from multiple directions rather than only from the conventional left and right channels. It requires spatial-audio-encoded content from a supported streaming service.

That should provide wider imaging and greater immersion than most television speakers, which are forced into increasingly thin cabinets and often struggle with basic dialogue intelligibility. Two physically separated speakers also have an inherent advantage over a compact soundbar trying to create stereo width from beneath the screen.

Does It Support Dolby Atmos?

This is where the reporting has become muddled.

The Google TV Streamer unquestionably supports Dolby Atmos. Google lists Dolby Atmos among its supported audio formats and promotes the device as capable of delivering immersive 3D sound when connected to compatible audio equipment.

Google also says two Home Speakers can create immersive Spatial Audio.

What Google does not currently say is that the Home Speaker pair constitutes an officially supported Dolby Atmos playback system.

Google’s setup documentation repeatedly uses the terms “Spatial Audio,” “spatial surround sound,” and “simulates surround sound.” It does not state that the speakers decode Dolby Atmos, identify them as a Dolby Atmos system, or describe how an Atmos soundtrack is rendered through the pair.

Some reports have combined the Google TV Streamer’s confirmed Atmos support with the speakers’ confirmed Spatial Audio feature and concluded that the complete pairing supports Dolby Atmos. That conclusion may eventually prove correct. The streamer could decode an Atmos soundtrack and render it through Google’s virtual spatial processing.

But Google has not documented that signal path clearly enough to present it as being true.

The accurate description is that the Google TV Streamer supports Dolby Atmos, while two paired Google Home Speakers reproduce compatible spatial content through Google’s Spatial Audio processing. Until Google explicitly confirms Dolby Atmos rendering through the speakers, calling the combination a Dolby Atmos system goes beyond the available documentation.

A logo on one box and spatial processing in two others do not automatically produce a certified Atmos system, regardless of how enthusiastically the dots are connected.

How Apple Handles the Same Question

Apple is considerably more explicit.

Apple states that one full-size HomePod or a stereo pair connected to an Apple TV 4K can automatically reproduce Dolby Atmos, Dolby Digital 7.1, and Dolby Digital 5.1 soundtracks. The full-size HomePod’s specifications also list Spatial Audio with Dolby Atmos for music and video.

HomePod mini can still be used alone or as a stereo pair for Apple TV audio, but it does not reproduce Dolby Atmos or Dolby surround formats. Those soundtracks are rendered through the smaller speakers in mono or stereo.

Apple also supports HDMI ARC and eARC through Apple TV 4K models from the second generation onward. That allows compatible televisions to send audio from game consoles, disc players, cable boxes, and internal TV apps through the Apple TV and onward to the HomePod system.

Google has not documented an equivalent television-audio return feature for its Home Speaker pairing. The confirmed Google configuration handles audio from the Google TV Streamer. There is currently no official indication that a PlayStation, Blu-ray player, or television tuner can route its sound through the speakers.

That is an important practical difference.

Did tvOS 26 Expand Apple’s Atmos Support?

Not significantly.

tvOS 26 allows any AirPlay-enabled speaker to be selected as the Apple TV’s default audio output. However, that does not turn every AirPlay speaker into a Dolby Atmos home-theater system. When third-party AirPlay speakers are selected, Apple notes that system alerts and game sounds continue to play through the television rather than the wireless speakers.

Apple’s full Dolby Atmos home-theater implementation remains limited to one or two full-size HomePods.

tvOS 18.5 previously added Wireless Audio Sync calibration for Dolby Atmos playback over AirPlay or Bluetooth. That feature corrects timing between picture and wireless audio; it does not add Atmos reproduction to speakers that lack it.

As of July 15, 2026, tvOS 26.5 is the latest public release. Apple describes it as a performance and stability update, while tvOS 26.4 corrected an audio issue when moving between Atmos and stereo programming. Neither update added rear HomePods, four-speaker configurations, Atmos support for HomePod mini, or Dolby Atmos playback through generic AirPlay speakers.

Better Than a Soundbar?

Google’s system has several appealing qualities. It is inexpensive, wireless, compact, and provides genuine stereo separation. The speakers also remain useful as Gemini assistants, Matter hubs, Thread devices, and components in a Google Cast multiroom system.

The limitations are equally obvious.

There is no dedicated center channel, subwoofer, rear channel, or height driver. Google has not published amplifier output, frequency response, or maximum volume specifications. Each speaker relies on one small full-range driver, and the surround presentation is created through processing rather than discrete speakers around the room.

A system such as the Klipsch The Sevens II costs considerably more but offers larger drivers, deeper bass, HDMI eARC, multiple physical inputs, a subwoofer output, and more convincing performance with both music and movies. A Dolby Atmos soundbar can also provide dedicated height processing and broader television connectivity.

Google’s system is not designed to replace either of those options. It is an affordable alternative to weak television speakers for people already invested in Google TV and Google Home.

That is still a worthwhile improvement when you consider the price.

The Bottom Line

Pairing two Google Home Speakers with Google TV Streamer gives Google something it should have delivered years ago: a simple and relatively inexpensive wireless television-audio system that connects its streaming, smart-home, and voice-assistant platforms.

The system should create a wider and more immersive presentation than most built-in television speakers, and $300 for the streamer and stereo pair is not unreasonable.

It should not yet be described as a confirmed Dolby Atmos speaker system.

Apple clearly states that the full-size HomePod and Apple TV 4K combination reproduces Dolby Atmos. Google currently says its two-speaker pairing provides Spatial Audio that simulates surround sound. Those statements are not interchangeable.

Google has finally found its way into the living room. It still needs to explain exactly what kind of soundtrack followed it through the door.

Where to buy:

Related Reading:

Forward-looking: Thirty years after Sega partnered with Nvidia to bring its arcade fighting game Virtua Fighter to PCs running on the Nvidia NV1 multimedia card, the two companies have once again teamed up to launch the franchise’s next title, Virtua Fighter Crossroads, on laptops and desktops powered by the Arm-based Nvidia RTX Spark. The collaboration will also extend to other upcoming Sega titles, the companies confirmed.

According to Nvidia’s press release, the new Sega games will support its latest upscaling and rendering technologies, including ray tracing and DLSS. The company added that its latest AI tools will also be supported by the upcoming titles, hinting at integration with its AI-powered neural rendering technologies, such as Reflex and G-Assist.

– VIRTUA FIGHTER CROSSROADS (@VFCROSSROADS) July 15, 2026

Neither Nvidia nor Sega revealed the names of the other games coming to RTX Spark, but online speculation suggests that several upcoming Sega titles, including Creative Assembly’s Alien: Isolation 2 and Total War: Warhammer 40,000, as well as the next installments in the Yakuza and Persona franchises, could receive support.

The announcements were made earlier today by Nvidia CEO Jensen Huang at an event in Tokyo celebrating the 30-year partnership between Sega and Nvidia. The gathering featured a reunion between Huang and former Sega president Shoichiro Irimajiri and was attended by Sega’s current CEO Haruki Satomi, COO Shuji Utsumi, and game designer Yu Suzuki.

Speaking at the event, Huang reminisced about how Irimajiri helped save Nvidia from bankruptcy in the 1990s by providing a $5 million cash infusion just as the company was running out of funding. Huang later admitted that Sega’s timely investment was the only thing that kept Nvidia afloat and that Irimajiri’s “understanding and generosity” saved the company from insolvency.

Sega had previously tapped Nvidia to develop a custom GPU for its Dreamcast console, but eventually canceled the order and used NEC’s PowerVR GPU instead, citing Nvidia’s outdated technology. Irimajiri still paid Nvidia the $5 million contract fee, but structured the payment as an equity investment, allowing Nvidia to develop the commercially successful Riva 128 and GeForce 256 gaming GPUs.

When Nvidia went public in 1999, Sega sold its shares for $15 million, tripling its original $5 million investment. While that may have seemed like a reasonable decision at the time, Nvidia’s market capitalization now sits at around $5 trillion, meaning Sega’s stake would have been worth hundreds of billions of dollars had the company held onto its shares.

Apple has now announced who can advertise on Apple Maps, barring entire business categories while still giving itself the right to final say over any advertisements on the platform.

On Tuesday, Apple updated its advertising guidelines. This was particularly noteworthy as it outlines the policies that advertisers will be expected to follow when advertising on Apple Maps.

Perhaps not surprisingly, Apple’s policies are generally more restrictive than Google’s across some key categories. This has typically been the case for content moderation between the two services.

For instance, Apple Maps has placed strict prohibitions on certain advertising categories outright. Banned categories include bail bonding services and cryptocurrency ATMs.

It also prohibits any home services. This would include plumbers, electricians, HVAC, roofing, and more; Apple does not explain the rationale for excluding these categories.

Google generally allows these categories to advertise, subject to local law and other advertising policies.

And while Apple hasn’t outright banned medical services, it does note that any such ad “will be evaluated on a case-by-case basis.” Google will generally allow medical ads, assuming they are compliant with local laws.

Apple has also fully banned political advertising, though this is not exclusive to Apple Maps. Google allows political advertising on Google Maps and other categories.

Apple Maps ads will be expected to comply with the whole of Apple advertising policy. This means that ads must be presented clearly and advertisers cannot send users to unrelated pages or use misleading claims.

Generally, Apple reserves broader discretion to reject ads, even if they’re not directly in violation of a rule. Google, on the other hand, tends to allow most things that don’t violate any of its listed criteria.

Coming to a Maps app near you

Apple Maps ads are only now beginning to roll out after years of speculation. Initially, it was believed that the tech giant was weighing the pros and cons of integrated ads in 2022.

Reports later suggested the advertisements would surface in 2023. However, the feature failed to materialize in 2023, 2024, or 2025.

Speculation resumed in late 2025, with reports once again hinting to a 2026 launch. Apple officially confirmed those plans in March 2026 with the launch of Apple Business, announcing that advertising would begin rolling out to Apple Maps during Summer 2026.

President Donald Trump restarted the Iran conflict with days of missile strikes, and US intelligence officials now estimate the total military cost of the war for the Pentagon could exceed $100 billion, according to two people directly familiar with the matter.

The officials were tracking the total cost of Operation Epic Fury to be in the $50 to $100 billion range at the end of May, dovetailing with confidential congressional estimates putting the costs to date at around $80 billion.

The Trump administration has not disclosed its cost estimates for the Iran war. In June the White House made a request for $88 billion to cover some of the costs of the war, but even that is an undercounting, the people say.

Part of the reason why a final cost is not available is that the Pentagon is still deciding whether to replace all the aircraft destroyed or damaged beyond repair during the conflict, the people say.

If the Pentagon decides not to replace certain aircraft, defense officials have told lawmakers, they will not request money for it—and therefore not factor that into the total cost of the war, the people say.

Presented with a detailed breakdown of this reporting, a War Department official told Inner Loop: “We have nothing further to announce at the moment.”

The nonpartisan Congressional Research Service said in a report on May 20 compiled using only publicly available reporting that the US had lost at least 17 manned aircraft and 25 drones since the start of the conflict.

The CRS report also showed the US had been losing an increasing number of drones, which are not cheap to replace. Among the 25 drones lost was an MQ-4C Triton, a high-altitude Navy surveillance aircraft that costs more than $600 million per airframe.

The cost of repairing US bases in the region, some of which sustained heavy damage from Iran firing retaliatory missiles and one-way attack drones in response to US strikes, will also be high.

Defense officials have told lawmakers behind closed doors they have not accounted for the costs of repair—and may never do so—if the US ultimately decides to shutter those bases because they are too vulnerable to Iranian attacks, the officials say.

Iran has been able to repeatedly hit several key bases in the Middle East in retaliatory strikes, including the headquarters of the US Navy’s Fifth Fleet at Naval Support Activity, Bahrain, which the Pentagon has not publicly acknowledged.

The only actual cost provided publicly by a top defense official has been from then acting Pentagon comptroller Jay Hurst, who testified in an oversight hearing in May that the cost of the war had risen to roughly $29 billion.

On Tuesday, at his nomination hearing to permanently become comptroller, Hurst declined to provide an updated figure but said the $29 billion was mainly munitions and the costs like fuel associated with having two US aircraft carriers steaming around the Middle East.

Operation “Gold Eagle” Has Arrived

The Trump administration on Tuesday launched a clearinghouse that will try to identify and patch any software vulnerabilities before malicious actors can hack them with the most powerful AI models.

An administration official told Inner Loop the clearinghouse, named “Gold Eagle,” will be run by the Cybersecurity and Infrastructure Security Agency, which will itself use AI models that are not publicly available, to identify vulnerabilities.

It marks the first major implementation of Trump’s June 2 executive order that aims to create a framework to oversee the rapidly growing threat of advanced AI models.

Edifier has expanded its MR Studio Monitor Series for 2026 with the new MR4.5, a near-field active loudspeaker designed for content creators, musicians, producers, and desktop listeners who want greater flexibility without moving into the larger and more expensive MR5.

Positioned between the MR4 and flagship MR5, the Edifier MR4.5 combines studio-focused sound reproduction with wireless connectivity, customizable tuning, and a compact design suitable for home studios, editing suites, gaming systems, and desktop audio setups.

The new model joins the MR3, MR4, and MR5 to complete Edifier’s current studio monitor lineup, offering buyers another option in a category where accuracy increasingly has to share desk space with convenience.

Related Reviews:

Features and Connectivity

Two Speakers: The Edifier MR4.5 includes two speakers. The active speaker serves as the system hub, handling connectivity, amplification, and signal distribution for both channels. No external amplifier, stereo receiver, or home theater receiver is required. The passive speaker connects to the active speaker using the supplied cable, which carries both power and audio signals.

Adaptable Setup: The MR4.5 includes acoustic tuning and desktop compensation controls that allow users to adjust its performance for different rooms, speaker positions, and listening distances. These controls can be useful when the speakers are placed on a desk, close to a wall, or in a small home studio where placement options may be limited.

Wired and Wireless Connectivity: The MR4.5 supports both wired connections and wireless streaming. Bluetooth 6.0 includes LDAC support for higher-resolution wireless audio, while multipoint connectivity allows the speakers to remain connected to more than one compatible device. This makes it easier to switch between a computer, smartphone, tablet, or other source without repeatedly disconnecting and pairing devices.

Bi-Amplified Speaker Design: The MR4.5 uses a bi-amplified design with an active crossover that divides the audio signal between the mid-bass driver and tweeter. Edifier designed the system to provide a balanced transition between the drivers and accurate sound reproduction for music production, video editing, gaming, and everyday desktop listening.

Extended Frequency Response: The MR4.5 has a specified frequency response of 50 Hz to 40 kHz. Edifier has tuned the system for a neutral, transparent presentation intended to reproduce recordings without excessive tonal coloration.

Dimpled Tweeter Waveguide: The MR4.5 uses a dimpled tweeter waveguide designed to improve high-frequency extension and clarity while creating a wider listening area. This should provide more consistent performance when the listener moves away from the central listening position.

80 Watts of Total Output: The MR4.5 delivers 80 watts RMS of total system output and a maximum sound pressure level of 98 dB at one meter. Although primarily intended for near-field listening, it should also provide sufficient output for some mid-field setups and medium-sized rooms.

Hi-Res Audio Support: The MR4.5 uses an analog front end and Class D amplifiers and supports digital audio signals up to 24-bit/96 kHz. Wireless playback is available through Bluetooth with LDAC support. Under compatible conditions, LDAC can transmit audio at up to 990 kbps and support signals up to 24-bit/96 kHz.

Wired Connection Options: Balanced XLR and TRS inputs allow the MR4.5 to connect to professional audio interfaces, mixers, and other studio equipment. RCA and 3.5 mm auxiliary inputs accommodate common consumer audio sources, while a headphone output supports private listening.

Bluetooth Multipoint: Bluetooth 6.0 provides a stable wireless connection, while Bluetooth Multipoint allows two devices to remain connected simultaneously. Users can switch between sources without disconnecting and re-pairing each time.

Room Acoustic Compensation: Rear knobs and the EDIFIER ConneX app provide multiple settings for room acoustic compensation, allowing you to adapt to different room modes and achieve the optimal listening experience.

Edifier MR Studio Monitors Comparison

The Bottom Line 

Although Edifier has enjoyed success with its MR Studio Monitor Series, a substantial gap remained between the MR4 and MR5 in both price and features. The MR4.5 fills that space by adding Bluetooth, LDAC, balanced XLR connectivity, room compensation, app-based adjustments, and an active bi-amplified crossover—features not offered by the more basic MR4.

The MR4.5 also provides a logical step down from the MR5, which uses a larger three-way driver configuration, a tri-amplified active crossover, and 110 watts of total output. Those additional capabilities may make the MR5 better suited to larger rooms and more demanding studio applications, but they may be unnecessary for desktop users and smaller production spaces.

The MR4.5 does not include HDMI ARC or eARC, USB audio, optical connectivity, or a phono input. Buyers who require HDMI eARC, USB-C, and optical inputs should consider Edifier’s M90 active speakers. The Onkyo Creator Series GX-30ARC goes further by combining HDMI ARC, USB-C, optical, and a line/phono input for direct connection to a turntable with a moving-magnet cartridge.

Price & Availability

Related Reading

For years, musicians have said AI song generators fed on their work without asking. A hacker just opened the black box and showed them exactly how.

Leaked source code from Suno, one of the biggest AI music tools, tells the story. It trained its model by scraping millions of songs and lyrics from across the internet. 404 Media got the data from a hacker who breached the company.

The haul is enormous. One file, labelled “youtube_music,” had logged more than 2 million clips. Others list tens of thousands of hours pulled from Deezer, Genius, and the stock library Pond5. In all, it adds up to decades of recorded music.

Ripping the vocals

The code is specific about what it wanted. To capture clean voices, it hunted for a cappella versions of songs on YouTube. To slip past YouTube’s defences, Suno routed its scraping through a proxy firm called Bright Data. It also trawled 420,000 podcasts, chasing roughly a million hours of speech.

The 💜 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

None of this is entirely new. In court, Suno has already admitted training on “essentially all music files of reasonable quality” on the open web. But the leak shows the machinery behind that line.

The record labels have long alleged as much. Suing Suno, the RIAA said the company copied “decades worth of the world’s most popular sound recordings,” and did it by “stream ripping” from YouTube, dodging the platform’s copy protections.

‘Fair use,’ Suno says

Suno’s defence, like most AI firms, is fair use. It says it trains on “publicly available music files” and builds its models for “original creation.” It even leaves artist names out of its training data, to discourage copycats.

On the breach, the company played it down. It called the November 2025 incident “limited” and “quickly contained,” said the exposed code was outdated, and insisted no sensitive data leaked. It does not hold customers’ full card numbers, it added, and it decided it did not need to tell users at all.

The hacker, who goes by ellie.191, tells a different story. They said they got in through the Shai-Hulud worm, grabbed an employee’s credentials, and pulled customer emails, phone numbers, and Stripe records. Suno never warned those users, some of them told 404 Media. The motive? “I like to hack anything and everything.”

The bigger fight

The timing matters. Some labels have already made peace, striking licensing deals with AI firms rather than fighting on. Sony is still in court, with a pivotal fair-use ruling due this summer.

Meanwhile artists keep saying the deals do little for them. Suno’s chief executive, Mikey Shulman, once said most people “don’t enjoy the majority of the time they spend making music.” The people whose music trained his model might disagree.