Let’s just get this out of the way: Mark Wahlberg is as A-list as it gets. He broke out in 1997’s “Boogie Nights,” and in the nearly three decades since, he’s built one of the highest-wattage careers in Hollywood: two Academy Award nominations, a Golden Globe nod for “The Fighter,” multiple Emmy nominations as a producer (HBO’s “Entourage” among them), and a filmography that spans crime dramas, comedies, and action blockbusters. Right now, he’s back on screen in “By Any Means,” a manhunt thriller set during the 1966 Mississippi civil rights killings.

He’s also signaling where he’s increasingly turning his attention. Wahlberg will sit down with Bruce K. Lee, founder and CEO of Keebeck Wealth Management, on the main stage at TechCrunch Disrupt 2026 to talk about that progression. Grab your ticket here so you don’t miss a moment of this session and Disrupt 2026. Ticket prices increase on September 25.

From Hollywood to business venture

His acting career is really just part of the picture. Over the past two decades, Wahlberg has built a production company, a restaurant chain, apparel and fitness ventures, a slate of angel investments, and the now 25-year-old Mark Wahlberg Youth Foundation, which supports inner-city kids and teens. He’s used his Hollywood success to launch big projects well outside the movie business — and in a fireside chat at this year’s Disrupt, he’s going to give us a peek into how.

Building an investing mindset

Wahlberg, who’s leaned almost entirely on entrepreneurial instinct about people, culture, products, and markets, is now someone who’s been building institutional-level investing discipline, with Lee as his guide, and with a growing focus on healthcare and wellness startups. It’s a rare, unvarnished look at how someone with Wahlberg’s cultural intelligence has worked his way into some of the most sophisticated rooms in business and finance.

That includes where high-profile entertainers and athletes sometimes get it wrong (Wahlberg is candid about having had to unlearn some of his own old instincts); how he’s used the trust built over a massive career to drive access and deal flow; and which technologies he’s most focused on right now and why.

Go inside Wahlberg’s playbook

If you’ve watched Wahlberg build brands, franchises, and businesses over the past decade and wondered, “Who is this guy?”—you won’t want to miss this conversation. Mistakes, mentorship, the mechanics of leveling up in a competitive investing landscape—we’ll cover it all at Disrupt.

Join Wahlberg and Lee on the Disrupt Stage, alongside 250+ of tech’s top leaders across 200+ sessions covering what it takes to build a viable startup and shape the next generation of innovation. On October 13-15, 10,000+ tech leaders will converge at San Francisco’s Moscone West. Don’t miss your chance to change the trajectory of your startup. Grab your pass now and save up to $200 before rates increase September 25.

There are two displays and multitasking options for iPhone Duo, but that doesn’t mean you’ll be able to play two videos at the same time.

If you’ve ever been scrolling a social media app while listening to music, you’ve likely had the frustrating experience of having the music pause randomly. That’s because the social media app took over as the primary audio provider.

The same effect occurs whenever you’re watching video on iPhone or iPad. Picture-in-picture, split view, or even dual screens won’t let a user play two videos at once, even on iPhone Duo.

While this may seem obvious to some readers, there was some question given that there are two distinctly different displays. Technically speaking, you could pair two sets of AirPods and play one video on the outer display and one on the inner display at the same time.

However, that use case isn’t available. There are some features, like in FaceTime, that utilize the multiple displays at once, but not for video.

There’s even a new multitasking view that lets users pin a video from an app like Apple TV to the top of the iPhone Duo while a second app remains active at the bottom. If the second app starts playing a video, the top video will pause.

The iPhone Duo still runs iOS, so it will have similar limitations to other iPhones in some cases. It will be available to pre-order on October 16 and ships on October 23.

With a fresh folding form factor, the new iPhone Duo is Apple’s most radical redesign in years. It lands alongside the more familiar-looking iPhone 18 Pro and iPhone 18 Pro Max. There’s no regular iPhone 18, or at least, not yet. Rumors suggest we may see a more affordable base model, perhaps flanked by an iPhone 18e and a second-generation iPhone Air, in spring 2027.

The two new Pro iPhones are available to preorder on September 12 and ship on September 18. You can preorder the iPhone Duo from October 16, ahead of a full release on October 23. But which iPhone 18 model is right for you? I’ll break down the differences and what these phones have in common, so you can choose with confidence. For options beyond the iPhone 18 lineup, read our Best iPhones guide.

Why You Might Not Need to Upgrade

Before you drop all those hard-earned dollars, remember that you probably don’t need the latest iPhone. If you have an older iPhone, Apple does a good job of updating the software (the iPhone 11 Pro is still supported), and for hardware issues, your iPhone might be repairable; maybe it needs a battery replacement or these troubleshooting tips. Don’t feel like you have to spring for Apple’s shiny new thing.

Sadly, Apple raised iPhone Pro prices this year, so both models are $100 more expensive than the iPhone 17 Pro series.

If you’re set on upgrading, consider raising funds by selling your old iPhone. Just be sure to back up your iPhone, and factory reset it before you sell or give it away. Once the new device arrives, we have tips on how to set up your new iPhone and what settings to change to get the most out of it.

Which iPhone 18 Is Right for You?

Photograph: Julian Chokkattu

Apple

iPhone Duo, iPhone 18 Pro, and iPhone 18 Pro Max

We’ll start by looking at what’s the same across the iPhone 18 models this year.

All three new iPhones have Apple’s A20 Pro chip inside, and, you guessed it, this is Apple’s most powerful chip ever. Now, iPhones are already the most capable phones for raw compute power, but the A20 Pro promises a significant jump in performance (around 20 percent faster and 30 percent more efficient than the last generation). It also packs a new neural engine to power on-device AI and a new thermal design with a larger vapor chamber to keep things cool. For the iPhone Duo, the A20 Pro has a display engine to ensure smooth transitions between two screens. While Apple doesn’t list the RAM, every model is rumored to have 12 GB.

While they come in different sizes and the Duo has two, every iPhone features an OLED screen that Apple calls a Super Retina XDR display. They also have ProMotion technology, otherwise known as a 120-Hz refresh rate. These adaptive refresh rates can drop as low as 1 Hz when there’s little activity on the screen, enabling features like the always-on display without draining the battery too much.

The displays are protected by the scratch-resistant and antireflective Ceramic Shield 2, the same as last year’s iPhone 17 range. And they can go from 1 nit up to 3,000 nits of brightness. Only the iPhone Duo has a custom nano-texture finish on the interior screen, giving it a matte look that’s designed to reduce glare and crease visibility.

All the phones have an IP68 rating for water and dust resistance, meaning they can survive a short dunk in fresh water. They all come in four storage options (256 GB, 512 GB, 1 TB, or 2 TB), support Qi2 magnetic wireless charging (MagSafe), USB-C 3.2, Dual SIM, Wi-Fi 7, Bluetooth 6, Thread, and Ultra-Wideband (UWB). It also has Siri AI, Apple Intelligence, Emergency SOS via Satellite, and Crash Detection.

To help you better understand which model to buy, here’s precisely where they diverge.

In brief: A Norwegian rescue helicopter and a small aircraft took off over Andøya in 2024, but the routes visible to satellite-navigation receivers told a different story. On tracking screens, the helicopter appeared to make a series of tight, overlapping turns. The plane seemed to trace sharp, unnatural patterns in the air. Neither display reflected the aircraft’s actual flight path.

The false tracks were created as part of Jammertest, an annual Norwegian exercise that examines how equipment responds when satellite-navigation signals are disrupted or manipulated.

Held on the island of Andøya, roughly 300 kilometers north of the Arctic Circle, the event brings together engineers, public agencies, and companies that depend on Global Navigation Satellite Systems, or GNSS.

The exercise, scheduled for mid-September this year, has become more relevant as real-world satellite interference has intensified around Russia’s war in Ukraine. An RAF aircraft encountered GPS jamming near the Russian border in May. In September 2025, Sweden’s Transport Agency said it was dealing with interference every day. Experts also warned in June 2026 that Russian satellites had transmitted interference from orbit, briefly disrupting GPS signals across Europe.

Aviation offers the clearest example of the problem because the effects are visible immediately. But GNSS interference can extend far beyond a flight deck or a map display. Satellite networks such as GPS provide not only location data but also highly accurate time. That timing supports communications networks, power grids, transport systems, and financial markets, where systems must remain synchronized.

Jammertest is built around the possibility that a receiver may lose that signal or, more troublingly, accept a false one. Jamming overwhelms the signal so that a receiver cannot use it. Spoofing sends a stronger counterfeit signal that can make equipment report an incorrect position or time.

The difference is important for infrastructure operators. A sudden loss of satellite data may be detected quickly, allowing a system to switch to backup timing. Spoofing can be harder to catch. If equipment continues to receive what appears to be valid data, it may distribute inaccurate timing or location information before anyone recognizes the problem.

At the 2024 event, Harald Hauglin, chief engineer of time and frequency metrology at the Norwegian Metrology Service, and his team demonstrated that spoofing can be a slow-moving risk. They gradually increased the strength of a spoofing signal over about 40 minutes. As the decoy signal became more powerful, a map display shifted the test site’s reported location away from its real position and into the Norwegian Sea.

The positional error was easy to see. Measuring the effect on time required a different setup. The team compared a GNSS-synchronized clock exposed to interference with a reference clock connected by fiber-optic cable to a receiver beyond the test area. The difference between the two was tracked in nanoseconds, revealing how a manipulated signal can slowly move a clock away from the correct time.

That is a particular concern for systems that need a common clock to operate. A power grid or mobile network can continue operating for some time after losing GNSS, but its ability to do so depends on the quality of its backup equipment. Local clocks, atomic references, and fiber-optic links can maintain precise timing, though building and maintaining them adds cost.

The more difficult scenario is a spoofing event that goes unnoticed. Rather than forcing a system to use its backup, it could continue feeding the system bad data. Over time, that may cause a wider synchronization problem across connected networks.

Jammertest is intended to identify these weaknesses before they appear in an actual disruption. Participants test commercial and industrial receivers in an open setting, then share results with other attendees. That approach differs from more tightly controlled US military tests, where information from the exercises is more restricted.

Norway and other governments are considering how much redundancy is needed as GNSS interference becomes a more regular feature of the security environment. Fiber-based timing networks and atomic clocks are among the options. Hauglin’s White Rabbit patch refers to White Rabbit, a CERN-developed system that distributes time across fiber-optic networks with accuracy better than one nanosecond.

Such systems can reduce dependence on satellite signals, but they do not replicate every service GNSS provides. Dana Goward, co-founder of the Resilient Timing and Navigation Foundation, said the answer will require several layers of protection rather than one replacement technology. “There is no silver bullet, which is why you need a combination of systems,” he told the BBC.

ai AND ml

GPT-Live-1 makes speaking to AI models more fluid

OpenAI has launched a version of its latest real-time full-duplex voice model as an API, allowing software developers to implement more responsive voice-enabled apps and conversational workflows.

GPT-Live-1 debuted in July, offering a way to interact through spoken prompts and responses instead of through typed text. Initially available through the ChatGPT interface, the voice model can now be reached via API calls.

As a full-duplex model, GPT-Live-1 can listen and generate speech at the same time, which aids fluid communication. People often talk over one another and may find it frustrating to take turns speaking and listening as if using half-duplex handheld radios.

OpenAI says it has focused on allowing developers implementing GPT-Live-1 to customize how users of business apps experience voice interaction and associated workflows.

“A core GPT‑Live‑1 strength, smooth interruption handling, is already delivering business impact: in early evaluations, Speak found that GPT‑Live‑1 gave learners more time to think before the language tutor responded, cutting interruptions by almost 80 percent versus previous turn-based systems,” the company said.

GPT-Live-1 is designed to handle spoken conversation while a backend model (eg, GPT-6 Astra or a more affordable option for rote tasks) handles information lookup, tool usage, and task management.

It represents a step up from OpenAI’s Realtime API, which debuted in 2024. In the Tau3 voice-agent intelligence benchmark, which assesses spoken customer-service tasks relevant to airlines, retailers, and telecom companies, GPT-Live-1 scores 86.2 percent, compared to 45.7 percent for GPT-Realtime-2.1 and 42.4 percent for GPT-Realtime-2.

Citing its own evaluations, OpenAI said GPT-Live-1 performs 30 percent better than GPT-Realtime-2.1 in Full Duplex Bench performance, while also outperforming its predecessor in turn-taking latency and interactive behavior.

Yelp has been using GPT-Live-1 for the Yelp Host service, and has nice things to say about the technology as it applies to the company’s AI restaurant reservation system. 

“With GPT-Live-1, every call is more conversational and responsive,” said Akhil Kuduvalli Ramesh, chief product officer at Yelp, in a statement. “Yelp Host now gives restaurants an advanced voice AI that delivers exceptional guest experiences, captures revenue opportunities they may have otherwise missed, and helps staff focus on serving guests instead of answering the phone.”

Yelp does not address whether restaurant customers making reservations welcome the more free-flowing robo-banter or simply accept it because it’s better than waiting on hold to chat with a person.

GPT-Live-1 costs $0.05 per minute of usage, on top of whatever backend model is running. Custom voices can be negotiated with the OpenAI sales team. The voice model is also available through OpenAI Presence, the company’s enterprise agent platform. ®