Sabrina Ortiz

Concerns over AI-generated content polluting scientific journal submissions have led many to question its role in research, but recent studies from MIT point to a more nuanced reality.

The use of AI in science raises questions on several levels. The most visible concern is "AI slop," the worry that AI-generated work is eroding research quality. A less discussed but equally important question is economic: even if AI is useful, does the time it saves scientists justify what it costs? MIT CSAIL’s paper AI Disruption and Falling Costs in Theory Research tackles exactly this question, and its findings are promising.

To determine the true value of these AI models, the researchers compare how models released from December 2024 to April 2026 solved cutting-edge theory questions in cases that have ground-truth human answers that were entirely unavailable to the AI. The answers came from 254 recent arXiv papers from seven scientific domains spanning mathematics, physics, and economics. The researchers reported that in those sixteen months, the changes in the models were "enormous":

• The share of problems for which models reached the same conclusion as the human authors (got full marks graded against the original research), rose from 3% (o1) to 35% (GPT-5.5) marking a 22 percentage point increase per year.
• These results were achieved at minimal cost, approximately $5 per attempt (or $15 per success) for GPT-5.5.
• Compute spending per full-marks answer fell from roughly $170 to $15, including failed attempts but excluding human labor and independent verification.
• A "back-of-the-envelope" calculation suggests that AI costs are less than 0.3% of what it would cost to have a human expert solve these problems.

Despite AI becoming this much more helpful, it doesn't mean its role in research is as simple as plug and play. The researchers observe that the value driven from the AI depends on the share of research costs and the savings achieved with them, citing an example of an AI model which in order to perform well would need extensive and costly training.

The researchers also draw attention to the fact that when some tasks become cheaper to do, the structure of science will need to evolve to take advantage of it. For instance, even when scientists save seven hours per week using AI, they have a growing backlog of untested hypotheses and substantial demand for output verification.

"Under these conditions, we would expect the field to evolve so that more scientists are focusing on these other complementary tasks," the researchers stated in the paper. "Based on our results, we would expect similar adjustments within the theory community, as some of their historical tasks become much cheaper to do with AI."

Our Deeper View

Some of the more exciting, useful, and benign use cases of AI involve breakthroughs in research. Yet there seems to be a lot of hesitancy about using it in this domain, because the ability to rapidly create more content with AI has led to watering down the quality of some research submissions. Arxiv wrote about this issue just this month as it has had to update its rate limit policy due to the influx of submissions, fueled by AI, and also, the increase in "dense, AI-written papers" described as "low-value." This MIT research presents the other side, in which AI could help fuel further breakthroughs by performing tasks that net out to be less expensive than the scientist doing the work themselves. The key here is that it doesn't replace the scientist, as they would still be the person checking the results, picking the right questions, and making the larger-picture connections (which they would have more time to do). It's also important to highlight that these models are extremely capable, as highlighted in the report or everyday breakthroughs. Just today, Anthropic published a report explaining how Claude designed protein binders against scorpion neurotoxins, part of a broader experiment to see how Claude can accelerate drug discovery.

Public sentiment toward the AI industry is broadly negative. Now, the companies are leaning into public-good initiatives to try and change the narrative.

On Sunday, Anthropic announced the Anthropic Public Benefit Fund, a new nonprofit arm that, subject to stockholder approval, will receive 5% of Anthropic's shares, with an additional 2% added at the board's discretion for every 50% increase in the company's value above the a set baseline, according to the blog post. Anthropic says the Fund is intentionally structured in a way that, the larger it gets, the more it can contribute to society.

While the company already operates as a public benefit corporation, or a company built to balance profit with public good, Anthropic says this new arm will have increased flexibility and scale to take actions such as making large donations to traditional nonprofits.

The fund will also work in tandem with the public good initiatives it already has in place, including:

• Committing $200 million to the Anthropic Economic Futures Research Fund that focuses on research of the economic impacts of AI;
• The companies' co-founders' pledge to donate 80% of their equity holdings;
• And the equity donation match program that allows employees to pledge equity which Anthropic matches. Half of its employees participate in it currently.

Meanwhile, Elon Musk, CEO and founder of SpaceX (and its AI arm known as SpaceXSI) announced on X that next year, he plans to give one of his SpaceX shares to "every kid in America." He also claimed that it would be the first of many, offering no additional details.

As CNBC reported, Musk currently owns about 4.8 billion SpaceX shares, and there were 72 million children under 18 in the US as of July 1, 2025, according to census data. Musk would be, therefore, giving away about $10.5 billion worth of shares, using SpaceX's closing stock price on Friday. As this announcement is limited to an X post, and Musk has a track-record for not following through on philanthropic promises, there is no guarantee that it will actually come to fruition. Yet, even if it does, it would still be lapped by Anthropic's contributions, which will likely come out to the $75 billion to $100 billion range, according to Reuters.

Our Deeper View

There is no denying that, at face value, these are generous donations whose outcomes would be beneficial, and certainly better than no contribution at all. It is also worth noting, however, what these companies stand to gain from taking actions of this sort. Prioritizing public perception is becoming increasingly necessary for them to carry out the plans they need to maximize profits. The data center buildout, for instance, depends in large part on local and federal policymakers, whose decisions are influenced by how they are perceived by the public. Without sustained goodwill, that support may not materialize. Securing favor with those in a position to influence those decisions matters just as much. Anthropic is in an especially vulnerable position: the company could go public as early as November, in what would potentially be the largest IPO of a public benefit corporation ever, and markets are moved by investor perception.

Though enterprises are eager to deploy AI, returns are still eluding many of them.

The Wharton School of the University of Pennsylvania has published its fourth annual report on the state of AI since the launch of ChatGPT, with the findings becoming one of the most closely watched bellwethers in the industry. Now, the report turns to the question everyone is asking: who is winning with AI, and what is their secret?

The survey, created in collaboration with market research and analytics firm GBK Collective, surveyed 1,416 US enterprise leaders and entry to mid-level employees, finding that there is a widening gap between organizations that simply implement AI and those who are seeing more meaningful transformations. The three markers of AI success it identified are proving value, transforming workflows, and aligning the organization around AI. However, only 7% of organizations surveyed succeed across all three metrics.

"Companies are now moving into the harder phase: integrating AI into how work gets done, learning new skills, changing processes, reorganizing departments. It’s like changing the engine while the car is running," Brian Smith, executive vice president of the AI, cloud and enterprise technology practice at GBK, told The Deep View. "That is hard work, and a lot less exciting than the early experimenting."

The report calls this year the fourth "wave" of AI, where the tech has been embedded in the core of enterprise operations, with 62% using the tech daily, 91% using it multiple times a week, and 73% reporting significantly changed workflows as a result.

Yet, the report identifies major flaws in adoption, including how companies are actually measuring ROI.

• A large portion of respondents, around 40%, reported judging return on investment by hours saved. Net-new revenue was the least utilized metric for success, with 27% reporting that they track this.
• However, the departments tracking returns by more than time saved are doing better comparatively: 35% of departments that track at least one growth measure report strongly positive returns, against just 13% of those tracking efficiency alone.
• Only 22% of managers look to AI to grow the business, whether through new revenue, new products, or better customer experience. Meanwhile, more than half are focused on improving productivity or cutting costs.

One way companies can implement more meaningful deployments of AI include having a designated Chief AI Officer position: 58% of managers with a newly designated CAIO report strongly positive ROI. Additionally, as enterprises lean into AI efficiency over tokenmaxxing, in tracking AI spend correlates with higher reported return.

Our Deeper View

Perhaps the most meaningful part of this report is seeing how the trajectory has changed over the past years. It allocates the first two "waves" of AI as the years 2023 to 2024 as exploration, marked by trial, error and experimentation by early adopters. Then, by wave three, which was 2025, regular usage of AI became more deeply integrated. Now, in wave four, as the report states that "adoption is no longer the question." Understanding this business mindset and why so many enterprises are willing to invest so much money so quickly also helps understand why the AI labs have redirected their attention to working professionals, creating two forces that are largely driving the AI market.

From smart wearables to AI-enhanced apps, using AI to log your days has become one of tech's hottest trends. Now Amazon is joining the race.

On Thursday, Amazon announced the Bee Frontier and Bee Eclipse, two personal AI gadgets that can listen to you throughout your days to provide summaries, to-do lists, reminders, and scannable directories of your day. While these devices are essentially a take on the AI pin offered by rival Plaud, Amazon is relying on its integration with Alexa as well as a lower price point to give it a competitive edge.

The Bee Frontier can be worn as a pin or as a wristband and was designed to be worn constantly, featuring seven days of battery life, IP67 water resistance, and two microphones with noise filtering that allows you to capture audio anywhere you go. This product is the next iteration of the Bee Pioneer, the startup's debut product that was also an AI note-taking wristband, before it was acquired by Amazon in July of 2022.

The Bee Eclipse serves the same purpose as the Bee Frontier with a different form factor: a slim, lightweight device that attaches to the back of your phone magnetically. It has more than 21 days of battery life, and a visible indicator when it is summarizing a conversation. While other competitors, such as Plaud, have offered the same products before, there are some key differentiators:

• Price: The Bee Frontier is $74.99 and the Bee Eclipse (available soon) is $99.99. Meanwhile, the Plaud Pin is $179 and the Plaud Note is $159.
• Alexa integration: With every purchase, users receive conversation summaries, recaps, and 14-day memory at no additional cost. However, with a Bee Pro subscription of $11.99, users get perks such as integrations with Amazon.com and Alexa+, which allow users to ask Alexa things such as "What are my Bee insights from yesterday?” or “What are my Bee to-dos?”

Of course, an always-on recorder like this raises obvious privacy concerns. Much like Apple’s approach to always-on listening with the S11 chip in the Apple Watch, Bee Private Compute uses a security architecture “that processes your conversations inside a secure and isolated processing environment where no one, including Amazon, can access your data,” according to the release. On top of that, audio isn’t stored, and users can turn off recording whenever they’d like and delete their transcripts.

Our Deeper View

When I first tried the Bee Pioneer in January 2025, what surprised me most was how genuinely useful its insights were. Being able to look back and reflect on the day’s interactions felt valuable on its own, and getting feedback on my actions was an added bonus. Still, two barriers have always held me back from devices like this: the upfront cost of hardware such as Plaud’s NotePin, and the siloed software they run on. If I want these kinds of in-depth insights, I don’t want to open yet another app or deal with a device that can’t connect seamlessly to the ecosystem I already use. That’s why the Alexa+ integration feels so useful. Many people already own an Echo speaker or smart display, and now they can simply ask it for these insights. Even something as simple as “When did my partner say he was coming home on Tuesday?” is helpful.

Apple is expanding its redesigned Siri assistant beyond the iPhone to home devices, the latest step in a years-long overhaul of the voice assistant.

On Tuesday, Apple launched a series of smart home products, with the star of the lineup being the new Apple HomeView, its first smart home display. While the display can do much of what competitors like the Amazon Echo Show and Google Nest Hub have done for years, including playing music, controlling smart home devices, and answering questions with AI, it leans much further into Apple’s signature ecosystem to offer something fresh.

For instance, the HomeView can facially recognize up to six users and then present each one with a personalized experience that displays the user's personal content from its Apple ecosystem of devices, such as Messages, FaceTime, Calendar, and Find My. The interactions with Siri are powered by Apple Intelligence and the context it has about you, making the responses personalized to the information stored within your Apple apps.

This was the most compelling part of the demos: Siri can now handle multistep requests, like “Hey, can you play the song my friend sent me?” It can also string together more complex tasks. For instance, in one demo, a user asked Siri when their friend would be home from the airport. Siri found the answer in a text message, worked out how long the drive home would take, and then helped the user schedule their coffee maker to turn on 15 minutes before the friend arrived.

The design also makes it unmistakably an Apple product, with subtle nods to the devices that came before it. The HomeView’s 7-inch Liquid Retina display swivels on a base that immediately reminded me of the 2002 iMac G4, and it comes in silver, starlight, rose pink, and space gray, the same colors of the iPhone 6s lineup. In classic Apple fashion, there’s careful attention to small details, like information that rearranges itself as you move closer to or farther from the screen.

There is also a separate wall-mounted HomeView model, available in space gray and silver, and both the version with a speaker at its base and the wall version retail for $299. It works with compatible smart home devices, as well as those from the recently announced Apple Home Collection, which include a new lineup of smart home cameras, thermostats, and a lock designed in collaboration with LG Electronics and Schneider Electric.

Our Deeper View

Apple has two very powerful advantages in the smart home market: People trust the company to keep their data private and safe, and Apple users are typically deep into the Apple walled garden and, therefore, are more inclined to use products within the ecosystem. Both of these things are typically the challenges I hear the most about when it comes to people who are opposed to having a smart home system. Users are fearful that it'll always be listening to them and are deterred by the fact that the setup isn't as seamless. Though I have not reviewed the device yet, if it's as easy to set up as it is for an Apple Watch or AirPods to pair with your phone, I could imagine this not only becoming a household product many Apple users want, but also a great gift with the holiday season around the corner.

Most large language models are built on human content, and newsrooms are paying the price: many readers now turn to AI for answers drawn from journalism that was never properly licensed.

As a result, the Standards for Publisher Usage Rights, or SPUR, coalition was formed in February 2026 to combat unauthorized AI content scraping. While it was first started by the UK-based publications, on Tuesday, the coalition welcomed more than 40 new members, including The New York Times, The Atlantic, The Observer, and The Economist, bringing the total up to 80 members and more than 20 countries globally.

Ultimately, the goal of SPUR is to bring the publishers together to set standards and rules for how AI systems can use publishers' copyrighted media, with licensing as the default way to do so. A key part of succeeding is showing a unified front, David Buttle, spokesperson at SPUR, told The Deep View.

"Over recent years, media publishers have been fighting the battle in silos for scant reward," said Buttle."That’s why the time is so ripe for this coalition. We now know this isn’t a battle that can be won with a fragmented front."

SPUR recently published the Content Telemetry Standard which is meant to create a standardized process in which AI labs can record which media is used to train their models and how they report that information to the publishers. It also says it is actively working with industry partners to make sure the framework is implemented effectively, and is developing tooling that makes it easy to do so.

The New York Times is a fitting partner for the coalition, as it has been one of the most outspoken publications regarding the misuse of AI using its content, suing OpenAI and Microsoft in 2023 on the basis of copyright infringement for using its content without consent to train its chatbots.

"High quality journalism is crucial to healthy democracies and to the functioning of AI products and experiences," Rebecca Grossman-Cohen, SVP of strategic partnerships and executive operations for The New York Times, said in the announcement. "We are proud to work with SPUR to develop robust standards and protections for copyrighted content that are not unilaterally set by tech platforms."

Our Deeper View

It’s really interesting how much the conversation around AI in newsrooms has shifted. Early on, many argued that the clearest way to protect journalism was to keep AI out of newsrooms entirely. Now, respected legacy outlets like Bloomberg and Reuters are using AI to improve reader experiences, such as summaries at the top of articles. This fits well with a line from the New York Times’ release: “Intellectual property protection and technological advancement can and must co-exist.” Ideally, newsrooms should be free to take advantage of AI, and even to decide whether to let AI companies use their content, if they find it beneficial. The real problem right now is that newsrooms often don’t have a say in those decisions at all.

Mark Zuckerberg has long promised that the future of wearable AI rests on a pair of glasses. The Ray-Ban Meta Audio gave me the most convincing look at that future yet.

The Ray-Ban Meta Audio are its first AI glasses with open-ear audio only, ditching the camera capabilities that the device has become known for. While forgoing the camera system may sound like a step backwards, the change gives it much more of an upgrade than you may think.

Everything else about the experience is mostly the same as previous generations, including two open-ear speakers and a 6-mic array for crisp listening and phone call experiences, touch and voice controls, and the Meta AI assistant. However, since the glasses don't have to make room for cameras, they are lighter and much more compact, weighing about eight to nine grams less than the Ray-Ban Meta Gen 3 and clocking in as the lightest pair in Meta's smart glasses lineup. As far as wearability, they felt no different than the prescription glasses I wear every day.

Sacrificing the camera comes with several other perks, including a retail price $100 lower than the Ray-Ban Meta Gen 3 at $349, and a longer battery life, with Meta claiming 12 hours from one charge (and an additional 48 from the charging case). In my testing, I couldn't kill the battery after wearing them for an entire day, with the biggest use case being playing audio while working or on walks.

However, I found the biggest perk to be not having the cameras in the frames during social interactions. Meta has faced backlash as a result of people feeling uncomfortable with Meta Ray-Bans wearers potentially recording or photographing them without their consent. Without the cameras, I feel much more emboldened to walk around, shedding the subconscious fear that I am making people uneasy.

For an average user who isn't a content creator, I don't think the lack of cameras will face pushback as users lean into the audio and voice features of these glasses. The collaboration also nailed the frames, Burbank and Clubmaster, the former resembling previous generations of Meta glasses, and the latter having the more unique and retro style.

Our Deeper View

Compared to previous iterations, these glasses feel the most wearable in social situations and in terms of overall comfort and practicality. However, the reason I have shared Zuckerberg's viewpoint is that the future of AI wearables lies in smart glasses because they can give AI the context of the world around you from your exact point of view. By removing the cameras, that benefit vanishes. While there are still significant benefits to getting rid of the cameras, the best of both worlds would be cameras that serve as sensors to feed context to the glasses, such as the Brilliant Labs Halo Glasses, which don't capture content, but rather use the visual feed as cues. Of course, the privacy problem would still exist, as users would have to trust Meta with feeds of their everyday life, but given how much information users are willing to give Meta Muse, I suspect it would be less of an issue.

Despite repeated calls for intervention, the US government has largely resisted AI regulation. Now, rogue agents may have pushed the limits too far.

Last week, Anthropic published a report detailing unintended actions by Claude observed in internal evaluations and use cases. Though the AI lab chose not to name the organizations to avoid exposing their vulnerabilities, it noted the incident involved "websites run by U.S. government agencies at the federal, state, and local levels." And according to a recent report from The New York Times, these incidents triggered the White House's most aggressive AI regulatory response yet.

Broadly speaking, Anthropic boiled Claude's actions down to four categories: exploiting a basic software flaw to run commands on a server, submitting a sensitive form on a real website, working around a restriction to access gated data, and using URL-shortening services to bypass limits. The New York Times talked to people familiar with the matter to fill in the gaps:

• Anthropic's agents submitted 20 visa applications via a form available on the State Department's website. The applications were incomplete and not processed.
• The lab's technology also submitted a false homicide tip to the Philadelphia Police Department via the agency's website, dated July 18. The tip was flagged as spam and not investigated.

The White House AI task force, now called the Super Intelligence Force, was formed the week Anthropic alerted authorities of the breach and was briefed on the agents' actions on Friday. The agency demanded that AI companies immediately report rogue AI activity, which, as The New York Times highlighted, marks the most aggressive AI regulation the Trump administration has implemented yet, as its attempts to regulate AI have thus far largely been voluntary and "morally binding."

“We informed the company that we expect immediate and full transparency to the entities involved and the public,” The agency said in a statement to The New York Times. “We also expect that the company will immediately provide remediation services to the affected entities and any harmed Americans.”

In the blog post, Anthropic says it identified most of the cases through a review of transcripts that began in July, and says that these cases have minimal real-world impact — less severe from an alignment and security perspective than the cybersecurity incidents it previously reported on July 30 and September 9. Yet, in the wake of these findings, the company decides to disable live internet access across all internal evaluations, not just high-risk and cybersecurity evaluations.

Our Deeper View

The cycle of AI evaluations leading to AI agents going rogue and attacking third-party websites illustrates why evaluations, red teaming, regulation, and other precautions are necessary. However, it also highlights the need for alignment on what risks are worth taking when testing AI, and a protocol for reporting them. While the Super Intelligence Force could be part of the answer, its formation seems to have been haphazard, with a Politico report stating that both Silicon Valley and Capitol Hill were caught off guard by President Donald Trump's announcement on social media. Though the requirement to immediately report rogue agent incidents could help address the problem, it remains to be seen how effective this task force will be in keeping up with these models' capabilities.

For years, augmented reality glasses promised more than they could deliver. Now, with AI unlocking a new set of capabilities, Meta is changing its headsets to get in the game.

Meta recently unveiled its Meta VR Glasses, which package what its traditional Meta Quest headsets can do into a significantly more compact form factor, abandoning the massive head-mounted contraption for something that looks much more like glasses. Meta also unveiled more natural ways to engage with the headset using gestures, which were the biggest standout in my demo.

To navigate the experience, you can also use native eye tracking. While this isn't unique to this device, it works very well and intuitively. Additionally, you don't even need to keep your hands in the frame, so with eye and hand gestures (index tap and thumb swipe), you can move and navigate the virtual experience conveniently. The VR/AR experiences are rich with detail, which makes sense given Meta’s deep experience in the space. One of my favorite Quest experiences, courtside basketball viewing, was excellent on these smaller glasses, likely thanks to their 5K Micro-OLED Infinite Display, which delivers 37 pixels per degree.

Comfort is where I have my qualms. At 100 grams, the glasses are far lighter than the 515-gram Meta Quest 2, but they’re front-heavy enough that they feel like they could slip off. I suspect that’s because the arms are as slim as regular sunglasses, and I’d gladly trade that look for a more even weight distribution. Portability is my other concern. I’d ideally wear these whenever I want a bigger screen on the go, but the pocket-sized computing puck gets in the way. Don’t get me wrong, it’s remarkable that Meta packed this many capabilities into a sunglasses form factor, but I may just be spoiled by Snap Specs.

The Snap Specs are a true facial computer with no pucks or additional accessories, and they have impressive graphics, too. While they weigh a bit more (132 grams for the 47mm frame size and 136 grams for the 52mm frame size) the weight is well distributed onto the arms. That does make them more bulky, but as said above, I think it is worth it to forgo the puck. However, the trade-off is that the Snap device costs $2,195, compared to $1,299 for Meta's version. Between these glasses and competitors like XREAL Aura, one thing is for sure: the lighter, more portable era of XR/AR is here.

Our Deeper View

AI is reshaping the AR/VR space in three important ways. First, it makes portable multi-screen setups more portable than ever. And since many professionals and developers now run multiple agents at once and need visibility into what each one is doing, these XR/AR glasses can offer that on the go. Second, AI makes apps far easier to build, which should lead to more immersive, worthwhile experiences and applications developed for the headsets. Finally, AI works best with context about what you’re seeing, doing, and hearing. Nearly all of these glasses already pack a host of sensors to track you and place you in space, and that same data can give your AI assistant the context it needs to be genuinely useful.