CONSUMER

OpenAI’s 6-device push reveals a frontrunner

You can't vibe code a lamp. 

OpenAI is racing headlong into becoming a hardware company, led by former Apple employees Jony Ive, Evans Hanke, Tang Tan, and others. Despite the Apple pedigree, the OpenAI hardware team appears to be operating much more like a Silicon Valley software or AI company by moving fast and trying a lot of different things.

A new report from The Information expanded the number of devices that OpenAI is prototyping. In fact, the report mentioned three new devices that I haven't seen reported anywhere else.

The list of reported devices from OpenAI now includes: 

• Smart glasses: These would presumably be similar to Meta Ray-Bans and forthcoming products from Google, Samsung, and Apple, but won't arrive until 2028. This was the most surprising of the three newly reported devices since Sam Altman had previously stated that the special device OpenAI was working on with Jony Ive was not a pair of glasses.

• Smart speaker: This is another of the newly reported devices and would compete with market-leading Amazon Echo speakers, which have been quite popular but are very limited in capabilities and are facing a five-year decline in sales. 

• Smart lamp: The last of the three newly reported devices, this would presumably be a smart speaker built into a lamp. Perhaps it might also tie into the recently announced ChatGPT Health and track your sleep.

• Smart earbud: More recently, reports have centered on OpenAI's Jony Ive device morphing into a smart earbud (or earbuds). The Foxconn supply chain leak on this one quoted a release date for this September. It still feels like the most useful and the most imminent.

• Smart pin: This smart pin sounds quite like the poorly received Humane AI pin. I suspect that OpenAI may have pivoted away from this device to the earbud mentioned above. However, Apple has also reportedly been working on an AI pin, so OpenAI might still be considering the form factor as well.

• Smart pen: Not to be left out, OpenAI has also reportedly been working on an AI smart pen. Like the earbud, this device has been tied to supply chain leaks connected to Foxconn. This one would presumably feed your handwritten notes in ChatGPT to add to voice and typed inputs. We can also imagine this device could have a microphone and/or camera to make it multimodal.

The biggest thing these OpenAI devices would have going for them — as much as the device aesthetics from the former Apple aficionados — is ChatGPT Voice (formerly Voice Mode). This feature isn't often discussed, but it's far more capable and usable than Siri or Alexa. I'd even give it a slight edge over Google's Ask Gemini (formerly Google Assistant), which is the only voice mode currently in its league.

Of all six of these reported devices, I still consider the AI smart earbud the most promising in the short term. It would likely be moderate in cost, it's a form factor most people already use, and even just having ChatGPT Voice at the ready would be a win. When we polled The Deep View audience earlier this year about buying an AI device from OpenAI, 43% said they would. So the brand still has work to do in winning over consumers. Also keep in mind that companies regularly leak reports like this to the press as trial balloons to gauge which devices draw the most buzz from consumers before deciding whether to actually manufacture them. So I predict we will not see all six of these devices come to market.

TOGETHER WITH ZED

Your next edit, already written

As you write code, edit predictions track your progress and offer the next edit before you ask for it.

Zed now supports multiple edit prediction models: Zeta, Zed's own purpose-built model; Mercury Coder, which uses a diffusion architecture to refine predictions in parallel; Sweep, trained with online reinforcement learning for suggestions under 100ms; and Ollama, to run open-weight models like Qwen or DeepSeek entirely on your hardware, with no code leaving your machine.

Which model performs best depends on your language and editing style, and now you can try them all in Zed!

PRODUCTS

Claude Code adds AI-powered security layer

Security has long been seen as one of AI’s biggest pitfalls, and the concerns have only accelerated as agents take on autonomous action. Anthropic is trying to change the narrative. 

On Friday, the company introduced Claude Code Security, an AI-powered tool that searches codebases for security vulnerabilities that humans may have missed, and it's now available in a limited research preview. 

Built into Claude Code, this tool weeds out security bugs and suggests software patches for human review.

• Rather than scanning for known patterns, the way traditional code scanning tools would, Anthropic claims that Claude Code Security reasons with your code by trying to understand how components interact and how data moves through your systems, “the way a human security researcher would.” 

• Before getting in front of human eyes, all of this tool’s findings then go through a multi-stage process to filter for false positives. Finally, everything is posted to the Claude Code Security dashboard for human approval. 

Claude Code Security addresses a prevailing issue for strapped security teams: “Too many software vulnerabilities and not enough people to address them,” Anthropic said. Finding subtle pitfalls “requires skilled human researchers, who are dealing with ever-expanding backlogs.” 

Anthropic’s tool adds another feature to its ever-popular suite of enterprise offerings. The feature comes as developers increasingly rely on Claude and other AI tools to do most, if not all, of their coding for them. 

But the more we rely on these tools, the more we risk when they fail. For example, a 13-hour Amazon Web Services outage in December reportedly involved the company’s own Kiro agentic AI coding tool, according to the Financial Times. Amazon blamed human error for this outage, claiming that the staffer involved had broader access permissions than expected and that "the same issue could occur with any developer tool or manual action.”

Trying to do AI the safe and ethical way is par for the course for Anthropic, yet this tool touches on a particularly poignant need. Developers across enterprises big and small are using Claude Code more than ever, and that reliance is only bound to grow as these companies search for ways to make their AI deployments gain traction. To trust the code they're writing, developers need to trust the codebases Claude is building on. Plus, a tool like this is bound to earn more reputation points for Anthropic if it can effectively prevent security snafus before they happen.

TOGETHER WITH CRUSOE

Skip the infrastructure headaches

Looking for a technically sound team to work with you to deploy your fine-tuned model on a platform built for performance?

Crusoe Managed Inference unlocks breakthrough speed and throughput without the infra overhead. You can move past the prototype phase with confidence and power your production apps with high reliability.

Your focus stays on building, not managing clusters—your fine-tuned models get best in class speed, throughput, and reliability without wrestling with infrastructure.

Our team is ready to optimize your fine-tuned model today.

STARTUPS

Perplexity shuns ads for enterprises, devices

As OpenAI jumps headfirst into advertising, another AI leader has taken a stance against it. We're not talking about Anthropic, we're talking about Perplexity. 

Last week, Perplexity told reporters at a press briefing that the AI search tool is backing away from its advertising plans and increasing its focus on its subscription business. Though it doesn’t intend to get rid of its free tier, Perplexity plans to pay for it by partnering with device makers moving forward. 

The decision follows OpenAI facing heat over its decision to embed ads into its popular chatbot. Perplexity executives fear that embedding ads into its platform may diminish consumer trust in its product. Instead, with its subscription business, Perplexity may be eyeing Anthropic’s primary audience: enterprises and developers. 

The move is a stark reversal from Perplexity’s previous strategy, in which CEO Aravind Srinivas said on a podcast last year that advertising will become its core stream of revenue over subscription and enterprise “if we crack it.” However, ads could have gone either way for the company:

• Given that Perplexity’s flagship product is an AI-powered search engine, leaning into advertising would have made sense, since ads are how Google makes its billions. 

• Still, where other ads-based businesses have Perplexity beat is on user numbers, with the AI search tool sporting roughly 60 million users compared to ChatGPT’s 800 million weekly active users, according to WIRED.

Perplexity doesn’t seem too worried about following in Google’s footsteps, telling reporters in the press briefing that “Google is changing to be like Perplexity more than Perplexity is trying to take on Google.”

There might be a reason attendees at the Cerebral Valley conference voted Perplexity a flop risk back in November: The startup does not have an easy road ahead of it to grow revenue. Though enterprise tech is generally a more lucrative bet than consumer, the startup has Anthropic to contend with, and its Claude Code and Cowork products have seen massive uptake in enterprise and developer circles due to their focus on trust and responsible AI. And among consumers, getting users to latch onto anything but its free tier is a hard sell, with Google providing its own AI search features and ChatGPT cementing itself as a household name. Perplexity will need to become known for something distinct in order to succeed.

LINKS

• General Catalyst commits $5 billion to India over the next five years

• Emerald AI raises $25 million to curb data center energy usage

• Scott Meyer, Snap’s head of AR glasses bet Specs, leaves the company

• Perplexity's Comet browser is on its way to the iPhone

• Sam Altman’s argument for AI energy use: Humans use a lot of energy, too

• Open source tools see an influx of low-quality AI-generated contributions

• Gemini 3.1 Pro: Google’s latest model, now smarter and more capable for complex problem-solving.

• Fern: An AI copilot for meetings that’s also an active participant, coaching users in real time. 

• Proximitty: AI-automated loan servicing and collections for banks and fintechs, no coding required. 

• Repaint: Use AI to build custom websites, just by chatting. This tool interviews you, imports your context and finds references to create your site rapidly.

• TikTok: Senior Research Engineer - Foundation Models, Ads Integrity

• Anthropic: Staff+ Software Security Engineer

• Tesla: Embedded Security Engineer, Vehicle Software

• Google: Research Scientist, Google Research, GenAI, Experiences

GAMES

A QUICK POLL BEFORE YOU GO