Connect with us

News

Tesla refines Full Self-Driving, latest update impresses where it last came up short

We were able to go out and test it pretty extensively on Saturday, and the changes Tesla made from the previous version were incredibly impressive, especially considering it seemed to excel where it last came up short.

Published

on

Credit: TESLARATI

Tesla released Full Self-Driving v14.2.1.25 on Friday night to Early Access Program (EAP) members. It came as a surprise, as it was paired with the release of the Holiday Update.

We were able to go out and test it pretty extensively on Saturday, and the changes Tesla made from the previous version were incredibly impressive, especially considering it seemed to excel where it last came up short.

Tesla supplements Holiday Update by sneaking in new Full Self-Driving version

With Tesla Full Self-Driving v14.2.1, there were some serious regressions. Speed Profiles were overtinkered with, causing some modes to behave in a strange manner. Hurry Mode was the most evident, as it refused to go more than 10 MPH over the speed limit on freeways.

Advertisement

It would routinely hold up traffic at this speed, and flipping it into Mad Max mode was sort of over the top. Hurry is what I use most frequently, and it had become somewhat unusable with v14.2.1.

It seemed as if Speed Profiles should be more associated with both passing and lane-changing frequency. Capping speeds does not help as it can impede the flow of traffic. When FSD travels at the speed of other traffic, it is much more effective and less disruptive.

With v14.2.1.25, there were three noticeable changes that improved its performance significantly: Speed Profile refinements, lane change confidence, and Speed Limit recognition.

Speed Profile Refinement

Speed Profiles have been significantly improved. Hurry Mode is no longer capped at 10 MPH over the speed limit and now travels with the flow of traffic. This is much more comfortable during highway operation, and I was not required to intervene at any point.

With v14.2.1, I was sometimes assisting it with lane changes, and felt it was in the wrong place at the wrong time more frequently than ever before.

Advertisement

However, this was one of the best-performing FSD versions in recent memory, and I really did not have any complaints on the highway. Speed, maneuvering, lane switching, routing, and aggressiveness were all perfect.

Lane Changes

v14.2.1 had a tendency to be a little more timid when changing lanes, which was sort of frustrating at times. When the car decides to change lanes and turn on its signal, it needs to pull the trigger and change lanes.

It also changed lanes at extremely unnecessary times, which was a real frustration.

There were no issues today on v14.2.1.25; lane changes were super confident, executed at the correct time, and in the correct fashion. It made good decisions on when to get into the right lane when proceeding toward its exit.

Advertisement

It was one of the first times in a while that I did not feel as if I needed to nudge it to change lanes. I was very impressed.

Speed Limit Recognition

So, this is a complex issue. With v14.2.1, there were many times when it would see a Speed Limit sign that was not meant for the car (one catered for tractor trailers, for example) or even a route sign, and it would incorrectly adjust the speed. It did this on the highway several times, mistaking a Route 30 sign for a 30 MPH sign, then beginning to decelerate from 55 MPH to 30 MPH on the highway.

This required an intervention. I also had an issue leaving a drive-thru Christmas lights display, where the owners of the private property had a 15 MPH sign posted nearly every 200 yards for about a mile and a half.

The car identified it as a 55 MPH sign and sped up significantly. This caused an intervention, and I had to drive manually.

Advertisement

It seems like FSD v14.2.1.25 is now less reliant on the signage (maybe because it was incorrectly labeling it) and more reliant on map data or the behavior of nearby traffic.

A good example was on the highway today: despite the car reading that Route 30 sign and the Speed Limit sign on the center screen reading 30 MPH, the car did not decelerate. It continued at the same speed, but I’m not sure if that’s because of traffic or map data:

A Lone Complaint

Tesla has said future updates will include parking improvements, and I’m really anxious for them, because parking is not great. I’ve had some real issues with it over the past couple of months.

Today was no different:

Advertisement

Full Self-Driving v14.2.1.25 is really a massive improvement over past versions, and it seems apparent that Tesla took its time with fixing the bugs, especially with highway operation on v14.2.1.

Advertisement

Joey has been a journalist covering electric mobility at TESLARATI since August 2019. In his spare time, Joey is playing golf, watching MMA, or cheering on any of his favorite sports teams, including the Baltimore Ravens and Orioles, Miami Heat, Washington Capitals, and Penn State Nittany Lions. You can get in touch with joey at joey@teslarati.com. He is also on X @KlenderJoey. If you're looking for great Tesla accessories, check out shop.teslarati.com

Advertisement
Comments

News

Tesla Semi is already winning over truck drivers

The consensus among participants is clear: the Semi feels quieter, quicker, and far less physically demanding than diesel rigs while delivering three times the power and dramatically lower operating costs.

Published

on

Credit: Tesla

Tesla’s all-electric Semi is proving more than just a flashy concept as it is winning converts among the professionals who know trucks best.

As fleets roll out Pilot Programs for Tesla across North America, drivers are raving about the Class 8 electric truck’s unique features, including a centered driver’s seat, massive touchscreen visibility, instant torque, and absence of gear-shifting fatigue.

These features are transforming long days behind the wheel into noticeably easier, less stressful shifts.

Tesla Semi pricing revealed after company uncovers trim levels

Advertisement

In a recent Wall Street Journal profile of early pilots, Dakota Shearer of IMC Logistics described backing out of a tight spot he had mistakenly entered:

“I backed right out of there, no problem. It’s like I’d never done it in the first place. That right there showed me that the technology the Tesla has makes a big difference.”

His colleague Angel Rodriguez of Hight Logistics, who switched from a 13-speed diesel, agreed:

“It’s just easier on your body. It’s less stressful because you’re not really having to engage the clutch and the stick shift.”

Advertisement

Veteran drivers in other tests echo the same enthusiasm. Tom Sterba, a Senior Driver at Saia, spent days testing the Semi and came away impressed with the navigation and overall feel:

“The navigation systems in these trucks are just unbelievable. That’s what I love about it.”

Sterba summed up the experience with a line that has since gone viral among trucking circles:

“I hope I retire in this truck.”

Advertisement

Pilot programs with ArcBest, thyssenkrupp Supply Chain Services, and Mone Transport delivered similar feedback. Drivers consistently praised the center-seat layout for eliminating blind spots, the smooth acceleration, and the overall comfort and safety.

Real-world data backed the hype, as ArcBest logged thousands of miles at efficient consumption rates, even over the challenging routes, like Donner Pass, while other fleets beat Tesla’s own efficiency targets.

The consensus among participants is clear: the Semi feels quieter, quicker, and far less physically demanding than diesel rigs while delivering three times the power and dramatically lower operating costs.

The latest chapter in the Semi’s story arrived just days ago on Jay Leno’s Garage, as Leno became the first outsider to drive the updated long-range production model, joined by Tesla Chief Designer Franz von Holzhausen, and Semi Program Director Dan Priestley.

Advertisement

Tesla reveals various improvements to the Semi in new piece with Jay Leno

The episode revealed major upgrades heading to volume production this year: the truck sheds roughly 1,000 pounds, adopts a 48-volt architecture, switches to fully electric steering with Cybertruck-derived actuators, and uses 4680 battery cells engineered for an over-one-million-mile lifespan.

Aerodynamics improved, enabling a 500-mile range on the long-haul version, and about 325 miles on the shorter-wheelbase standard-range model. Megachargers can now deliver up to 1.2 megawatts, adding roughly 300 miles in about 30 minutes.

Leno hauled heavy loads and marveled at the turning radius and effortless power delivery. “I don’t feel like I’m pulling anything,” he said during the episode.

Advertisement

With hundreds of Semis already accumulating over 13.5 million fleet miles and high uptime, the future of heavy-duty trucking looks electric. Drivers are giving raving reviews, and they’re ready to climb aboard the electric trucking industry for good.

Continue Reading

Investor's Corner

Tesla and SpaceX to merge in 2027, Wall Street analyst predicts

The move, Ives argues, is no longer a distant possibility but a logical next step, fueled by deepening operational ties, shared AI ambitions, and Elon Musk’s vision for dominating the next era of technology.

Published

on

Credit: Grok

Tesla and SpaceX are two of Elon Musk’s most popular and notable companies, but a new note from one Wall Street analyst claims the two companies will become one sometime next year, as 2027 could see the dawn of a new horizon.

In a bold new research note, Wedbush analyst Dan Ives has reaffirmed his long-standing prediction: Tesla and SpaceX will merge in 2027.

The move, Ives argues, is no longer a distant possibility but a logical next step, fueled by deepening operational ties, shared AI ambitions, and Elon Musk’s vision for dominating the next era of technology.

He writes:

Advertisement

“Still Expect Tesla and SpaceX to Merge in 2027. We continue to believe that SpaceX and Tesla will eventually merge into one company in 2027 with the groundwork already in place for both operations to become one organization. Tesla already owns a stake in SpaceX after the company’s $2 billion investment in xAI got converted to SpaceX shares following SpaceX’s acquisition of xAI earlier this year initially tying both of Musk’s ventures closer together but still represents <1% of SpaceX’s expected valuation. The recent announcement of a joint Terafab facility between SpaceX and Tesla further ties both operations together making it more feasible to merge operations given the now existing overlap being built out across the two with this the first step.”

The groundwork is already being laid. Earlier this year, SpaceX acquired xAI, converting Tesla’s $2 billion investment in the AI startup into a small equity stake, less than 1 percent, in SpaceX.

Regulatory filings cleared the transaction in March 2026, formally linking the two Musk-led companies financially for the first time. Then came the announcement of a joint TERAFAB facility in Austin, Texas: two advanced chip factories, one dedicated to Tesla’s AI needs for vehicles and Optimus robots, the other targeting space-based data centers.

Elon Musk launches TERAFAB: The $25B Tesla-SpaceXAI chip factory that will rewire the AI industry

Advertisement

Ives calls Terafab the “first step” toward full operational integration.

SpaceX’s impending IPO, expected as soon as mid-June 2026, will turbocharge these plans. The company aims to raise approximately $75 billion at a roughly $1.75 trillion valuation, far exceeding earlier estimates.

Proceeds will fund Starship rocket flights, a NASA-contracted lunar base, expanded Starlink services across maritime, aviation, and direct-to-mobile applications, and crucially, orbital AI infrastructure

A major driver is the exploding demand for AI compute. U.S. data centers are projected to consume 470 TWh of electricity by 2030, constrained by power grids and land.

Advertisement

SpaceX’s strategy, launching millions of solar-powered satellites to host data centers in orbit, bypasses Earth’s energy bottlenecks. Solar energy captured in space avoids atmospheric losses and day-night cycles, offering a scalable solution for AI training and inference.

Advertisement

The xAI acquisition ties directly into this vision, positioning the combined entity as a leader in extraterrestrial computing.

The merger would create a formidable conglomerate spanning electric vehicles, robotics, satellite communications, human spaceflight, and defense.

Ives highlights SpaceX’s role in the Trump administration’s “Golden Dome” missile defense shield, which would leverage Starlink satellites for tracking.

For Tesla, access to SpaceX’s launch cadence and orbital assets could accelerate autonomous driving, Robotaxi fleets, and Optimus deployment.

Advertisement

Musk, who has signaled his desire to own roughly 25 percent of Tesla to steer its AI future, views the combination as essential to overcoming fragmented regulatory scrutiny from the FTC and DOJ.

Challenges remain. Antitrust hurdles could delay or reshape the deal, and shareholder approvals on both sides would be required. Yet Ives remains bullish, maintaining an Outperform rating on Tesla with a $600 price target, implying substantial upside from current levels. The analyst sees the merger as the “holy grail” for consolidating Musk’s disruptive tech empire.

If realized, a 2027 Tesla-SpaceX union would not only reshape corporate boundaries but redefine humanity’s trajectory in AI and space exploration. It would mark the moment two pioneering companies become one unstoppable force, pushing the limits of what’s possible on Earth and beyond.

Advertisement
Continue Reading

News

Tesla ‘Killer’ heads to the graveyard as AFEELA taps out

SHM has officially discontinued development of its highly anticipated AFEELA electric vehicles. On March 25, the joint venture between Sony and Honda announced it would halt the AFEELA 1 luxury sedan and a planned SUV model.

Published

on

Credit: AFEELA/X

There have been many Tesla “Killers” over the years, all of which have either failed to dethrone the automaker from its dominance in the United States, or even make it to the market altogether.

The Sony Honda Mobility (SHM) project, known as AFEELA, is the latest to make it to the grave, as the company announced its intentions to abandon the project earlier this week, Bloomberg reported.

SHM has officially discontinued development of its highly anticipated AFEELA electric vehicles. On March 25, the joint venture between Sony and Honda announced it would halt the AFEELA 1 luxury sedan and a planned SUV model.

The decision follows Honda’s March 12 reassessment of its electrification strategy, which scrapped several upcoming EV programs amid slowing demand, high costs, and shifting market conditions.

Advertisement

SHM stated that it could no longer rely on key Honda technologies and manufacturing assets, leaving “no viable path forward.” Reservation fees for early buyers in California are being fully refunded, and the joint venture’s future is now under review.

Launched with fanfare in 2022, the AFEELA was positioned as a tech-forward premium EV blending Honda’s engineering reliability with Sony’s entertainment and AI expertise.

Prototypes featured advanced autonomous driving systems, immersive in-cabin displays, and even PlayStation integration, earning it early media labels as a potential “Tesla Killer.”

No more “Tesla Killers:” It’s becoming increasingly difficult to distinguish the “EV market” from the mainstream auto segment

Advertisement

Priced around $90,000, the sedan was slated for limited production at Honda’s Ohio plant with deliveries targeted for late 2026. Industry watchers saw it as a serious challenger to Tesla’s dominance in software, connectivity, and premium appeal.

Yet, like many ambitious EV projects, it fell victim to broader industry headwinds: softening consumer demand, persistent high interest rates, and intense competition from established players.

The AFEELA joins a long list of vehicles once hyped as “Tesla Killers” that failed to deliver. In the late 2010s, Fisker’s second act, the Ocean SUV, promised stylish design and solid-state battery tech but collapsed into bankruptcy in 2024 after production delays, quality issues, and financial shortfalls.

Faraday Future poured billions into the FF 91 luxury sedan, touting it as a hyper-tech rival with unmatched performance and features; the company delivered fewer than 100 vehicles before fading into obscurity.

Advertisement

Lordstown Motors’ Endurance electric pickup generated massive pre-order buzz and Wall Street excitement but imploded after exaggerated range claims, a factory sale, and eventual bankruptcy.

Even Lucid Motors’ Air sedan, frequently called a Tesla slayer for its superior range and luxury, has struggled with sluggish sales and missed growth targets despite strong reviews.

Lucid unveils Lunar Robotaxi in bid to challenge Tesla’s Cybercab in the autonomous ride hailing race

Rivian’s R1T and R1S trucks enjoyed similar early acclaim and a blockbuster IPO, yet production ramp-up challenges and profitability woes have prevented it from dethroning Tesla.

Advertisement

The AFEELA’s quiet demise underscores a harsh reality in the EV sector. While Tesla’s first-mover advantage in software, charging infrastructure, and brand loyalty remains formidable, legacy automakers and tech newcomers alike continue to underestimate the complexities of scaling affordable, desirable electric vehicles.

As market realities force tough choices, the graveyard of “Tesla Killers” grows longer, another reminder that innovation alone is rarely enough to topple an established leader.

Continue Reading