Tesla’s Robotaxis Hit the Street — Regulators Struck Back Fast

Federal regulators opened an investigation into Tesla’s driverless Cybercab within hours of its commercial launch. The vehicles hit the streets of Austin, Texas, for the first time this week. The Tesla Cybercab safety probe centers on whether Tesla was right to certify the vehicle as meeting federal safety standards. The Cybercab lacks a steering wheel, brake pedal or mirrors, according to the National Highway Traffic Safety Administration.

What the Tesla Cybercab safety probe is actually examining

NHTSA opened what is known as an Audit Query. That is a formal review process. It lets the agency examine the process and technical data behind a manufacturer’s self-certification decision. Tesla told NHTSA it believes the Cybercab complies with all applicable federal vehicle safety standards. The agency will test whether Tesla was right to decide that certain standards, including those requiring manual controls, simply do not apply to a vehicle built without them.

Tesla Cybercab safety probe
Cybercab’s lack of manual controls is the central question in NHTSA’s new safety review.

TechCrunch reported the vehicles lack permanently attached brake and gas pedals by design. They also lack a steering wheel and rearview mirrors. That is a break from every prior commercially deployed robotaxi in the US. Earlier vehicles kept manual controls even when running autonomously.

Why regulators moved so fast

NHTSA opened its probe within hours, not weeks, of Cybercab’s Austin launch. That speed reflects how unusual Tesla’s self-certification claim really is. Most automakers deploying autonomous vehicles keep manual controls in place as a fallback. That serves two purposes: regulatory comfort, and letting a safety driver or remote operator intervene using familiar controls. Tesla removed them outright. That puts a core legal question directly in front of federal regulators for the first time at commercial scale: can a vehicle without manual controls meet standards written assuming those controls exist?

What has already gone wrong on the road

Beyond the certification question, Tesla’s broader robotaxi pilot in Austin has already drawn scrutiny. Erratic driving incidents were caught on camera on public highways, according to earlier CBS News reporting. Separately, the National Transportation Safety Board has examined why Tesla has been slow to report crashes involving its self-driving systems in the past. That history adds context for why regulators are treating this launch with particular caution.

What NHTSA’s probe means for Cybercab’s future

An Audit Query does not halt Cybercab service on its own. But it can lead somewhere more serious. NHTSA could order a formal defect investigation, a recall, or a requirement that Tesla add back some of the removed controls if it finds the self-certification was flawed. Audit queries typically run for months, not days. Cybercab will likely keep operating in Austin while the review proceeds, unless a specific safety incident forces faster action.

Other automakers developing driverless taxi services are watching closely, since a finding against Tesla’s approach could reshape how the entire industry designs vehicles without manual controls.

How other cities are watching this test

Austin is not the only city where driverless taxis operate. Several other US cities host similar pilots from Tesla and its rivals. None of them, so far, has approved a vehicle built entirely without manual controls. Regulators and city officials elsewhere have a direct stake in how NHTSA’s review turns out, since it will shape whether similar no-pedal designs get a path to market in their own jurisdictions.

That makes this case bigger than one company or one city. If Tesla’s approach survives NHTSA’s review, other manufacturers may follow with their own no-pedal, no-wheel designs. If it does not, the entire industry may need to keep manual backups in future robotaxi models for years to come.

Insurance companies underwriting these vehicles are also watching closely. A finding that Tesla’s certification process was flawed could complicate coverage terms for Cybercab and any future vehicle built on the same design philosophy, regardless of how the underlying safety record actually looks on the road.

A handful of individual ride accounts, good or bad, cannot replace systematic safety data. That is precisely why an Audit Query exists. It looks at the underlying engineering and testing record, not anecdotes from early riders.

Tesla has weeks, not days, to prepare its response to NHTSA’s questions in full. How thoroughly it documents its certification reasoning will likely determine whether this review closes quietly or escalates into something with real consequences for the Cybercab program.

Need to know

What is NHTSA investigating about the Tesla Cybercab?

NHTSA opened an Audit Query into whether Tesla correctly certified that the Cybercab meets federal motor vehicle safety standards despite lacking a steering wheel, brake pedal and mirrors.

When did the Cybercab launch in Austin?

Tesla put the first commercial driverless Cybercabs on Austin streets in early September 2026, with NHTSA opening its probe within hours.

Does the Cybercab have a steering wheel or pedals?

No. The vehicle is built without permanently attached manual controls, including a steering wheel, brake pedal, gas pedal and rearview mirrors.

Has this stopped Cybercab from operating?

No. An Audit Query does not automatically halt service, though it could lead to a formal investigation or recall if NHTSA finds problems.

Has Tesla faced scrutiny over self-driving before?

Yes. The National Transportation Safety Board has previously examined delays in Tesla’s crash reporting involving its self-driving technology.

Sources

  • NHTSA — Investigation into Tesla Cybercab Self-Certification Following Austin Deployment. nhtsa.gov
  • TechCrunch — Feds launch investigation into Tesla’s Cybercab deployment. techcrunch.com

Related coverage on Tamara News: the EU’s AI Act transparency rules and Apple’s UK antitrust lawsuit.

OpenAI Just Released Its Most Powerful Model Yet — With a Catch

OpenAI began rolling out its newest flagship model on September 3. The company calls it the most capable system it has ever shipped. The GPT-6 Astra release arrived months later than originally planned. OpenAI said it added extra safeguards after a security incident earlier in the year, according to CNBC.

What the GPT-6 Astra release actually includes

Astra ships with a one-million-token context window. Its API price runs roughly 2.5 times higher than its predecessor, GPT-5.6 Sol. OpenAI set API access at $10 per million input tokens and $50 per million output tokens. Cached input, batch processing and a slower “flex” mode all cost less. The rollout started with a limited group of trusted partner organizations. A wider release to ChatGPT’s paid tiers, the API, and cloud partners including AWS follows in the coming days.

GPT-6 Astra release
Frontier AI models now ship with benchmark claims that outside researchers have not yet independently verified.

OpenAI’s own benchmark claims are striking. The company says Astra scores 98% on FrontierMath Tier 4. It also claims 99.9% on ARC-AGI-3 and 100% on ExploitBench, a security-focused benchmark. Those numbers come from OpenAI itself. They are not independently verified, and outside researchers have not yet had time to reproduce them at scale.

Why the release was delayed

OpenAI said it pushed back Astra’s launch after a breach linked to its Hugging Face-related infrastructure in July. The company used the extra time to add safeguards. It says those safeguards are meant to reduce the risk of serious misuse. Al Jazeera reported the launch is drawing heightened scrutiny. Astra’s advanced cybersecurity capabilities are a selling point, but they also raise questions about dual-use risk.

What OpenAI says about the safety tradeoffs

OpenAI said this week it believes the added safeguards “sufficiently minimize the risk of severe harm” ahead of a broader release. That is the company’s own assessment of its own product. It is not an independent audit. It is worth watching closely as more researchers and enterprise customers get hands-on access in the coming weeks.

What Astra’s rollout looks like from here

The next phase brings Astra to ChatGPT’s paid subscribers and general API access. At that point, far more developers and businesses can test its claimed capabilities against real workloads. That is a very different test than OpenAI’s own benchmark numbers. Enterprise security teams evaluating Astra’s cybersecurity features will likely publish independent findings first, given how directly those capabilities intersect with existing security tooling.

Competing labs are expected to respond within weeks with their own updated models, a pattern that has held throughout 2026 as each major release resets the competitive bar.

How rivals are likely to respond

OpenAI does not operate in a vacuum. Google, Anthropic and several well-funded startups have each shipped major model updates this year. Every time one lab leaps ahead on benchmarks, the others tend to respond within weeks rather than months. That pattern held throughout 2026, and there is little reason to expect it to break now.

What makes Astra different is the cybersecurity angle. Most prior releases competed mainly on reasoning and coding benchmarks. Astra’s marketed strength in exploit-related tasks puts security teams, not just developers, directly into the conversation about what a frontier model should be allowed to do.

Enterprise buyers evaluating Astra will likely ask pointed questions about access controls and misuse monitoring before deploying it widely. Vendors that sell security software are watching this rollout closely, since a genuinely capable exploit-finding model changes the calculus for both attackers and defenders.

A model marketed for its offensive security strength is exactly the kind of release that tends to draw regulatory attention in both the EU and UK, given each region’s growing focus on frontier AI oversight. How that scrutiny plays out here is likely to shape how future releases in this category get framed.

For now, most of what the public knows about Astra’s real-world risk profile comes from OpenAI’s own summary rather than an outside, independent review. That gap tends to narrow over the weeks after a major model launch, as more outside researchers get access and publish their own findings.

That review process, more than any single benchmark number, will ultimately decide how Astra gets remembered: as a genuine leap forward, or as a launch whose claims outran what the model could reliably deliver once wider testing began.

What people are asking

When did GPT-6 Astra launch?

OpenAI began rolling out GPT-6 Astra on September 3, 2026, starting with a limited preview for trusted partner organizations.

How much does GPT-6 Astra cost to use via the API?

OpenAI priced Astra at $10 per million input tokens and $50 per million output tokens, roughly 2.5 times the price of its predecessor, GPT-5.6 Sol.

Why was the release delayed?

OpenAI said it delayed the launch to add safeguards following a security incident linked to its Hugging Face-related infrastructure in July.

What is Astra’s context window?

Astra supports a one-million-token context window, according to OpenAI’s release details.

Have Astra’s benchmark scores been independently verified?

Not yet. The reported scores, including 98% on FrontierMath Tier 4 and 100% on ExploitBench, come from OpenAI’s own testing.

Sources

  • CNBC — OpenAI announces rollout of GPT-6 Astra model. cnbc.com
  • Al Jazeera — OpenAI unveils GPT-6 Astra amid rising scrutiny and safety concerns. aljazeera.com

Related coverage on Tamara News: the AI chatbot outage earlier this month and ChatGPT’s DSA designation in the EU.

Broadcom’s Numbers Were Great — Except the One That Mattered Most

Broadcom shares fell more than 2.5% in extended trading this week. The chipmaker issued a fourth-quarter revenue outlook that landed just below what Wall Street expected. The Broadcom Q4 revenue forecast called for roughly $34.8 billion. Analysts had averaged estimates near $35.03 billion, according to the Globe and Mail.

Why the Broadcom Q4 revenue forecast disappointed investors

The shortfall was small in percentage terms. But it arrived alongside signals of intensifying competition in custom AI processors. That segment is what investors have leaned on most to justify Broadcom’s valuation. Investing.com reported the soft guidance overshadowed otherwise strong growth elsewhere in the business. That pattern has repeated across several AI-linked chip names this earnings season.

Broadcom Q4 revenue forecast
Custom AI processors have become the fastest-growing and most closely watched part of Broadcom’s business.

Broadcom’s own numbers were not weak in isolation. Fourth-quarter consolidated revenue is still projected to grow 93% year over year to that $34.8 billion figure. The company also expects to hold its non-GAAP operating margin at 66%, a level most industrial companies would consider exceptional.

The AI chip business is still growing fast

Broadcom expects AI semiconductor revenue to accelerate to $21.7 billion in the fourth quarter. That is up 236% year over year. This growth rate outpaces the company’s overall revenue growth by a wide margin. It underscores how central custom AI chips have become to Broadcom’s near-term story, even as headline guidance missed estimates.

What “below estimates” actually means here

A guidance miss of roughly $230 million on a $34.8 billion forecast is a rounding error in absolute terms. But stock prices react to the gap between expectation and reality, not the underlying scale. Analysts had priced in near-flawless execution from Broadcom’s custom silicon business. Any signal of rising competition was always going to move the stock more than the raw numbers suggest.

The quarter ahead

Broadcom’s next earnings report will answer a key question. Does the softer guidance reflect a one-quarter blip from customer order timing? Or is it an early sign that rivals are winning a larger share of custom AI chip contracts? Investors will watch commentary on customer concentration closely. Broadcom’s custom processor business depends on a small number of very large cloud computing customers rather than a broad base.

Semiconductor peers reporting later this month will offer an early read on whether Broadcom’s guidance miss reflects an industry-wide slowdown in custom chip orders or a company-specific issue.

Why analysts prize this one number

Wall Street has treated Broadcom’s custom AI chip guidance as a proxy for the entire custom silicon industry this earnings season. That is why a modest miss triggered an outsized stock reaction. Investors are not just pricing Broadcom’s own results. They are using this number to guess how fast big cloud providers plan to keep spending on custom processors instead of buying off-the-shelf chips from rivals.

Several major cloud customers design chips jointly with Broadcom under multi-year agreements. Any sign those customers are slowing orders, even slightly, ripples through the stock faster than the dollar figures alone would suggest. That is the real story behind Wednesday’s drop.

By most historical standards, this remains a strong quarter for Broadcom. Investors, for now, appear more focused on the trend line in custom chip orders than on the overall scale of the business.

Large options positions often build up ahead of major chip earnings. That kind of positioning can amplify price swings in either direction once results land, which helps explain why a relatively modest guidance gap can produce an outsized move in a stock like Broadcom’s.

Broadcom’s next report will land after a full quarter of the AI chip market’s current competitive dynamics play out. That report, more than this week’s reaction, will show whether Wednesday’s guidance was conservative positioning or an early signal of a genuine slowdown.

Rival chipmakers reporting in the meantime will offer an earlier read on whether the caution is industry-wide or specific to Broadcom’s customer mix. Investors will parse those results closely for hints before Broadcom reports again.

Until then, Wednesday’s reaction stands as a reminder of how thin the margin for error has become at the top of the AI chip trade. Even a strong quarter is not enough on its own if guidance falls short of a very high bar.

That bar keeps rising each quarter, set less by Broadcom itself than by how aggressively the broader AI infrastructure buildout keeps accelerating around it.

Fast facts and FAQs

Why did Broadcom stock fall after earnings?

Broadcom issued fourth-quarter revenue guidance of about $34.8 billion. That was below the roughly $35.03 billion analysts expected, sending shares down more than 2.5% in extended trading.

Is Broadcom’s AI chip business still growing?

Yes. The company expects AI semiconductor revenue to reach $21.7 billion in the fourth quarter, up 236% from a year earlier.

What is Broadcom’s expected profit margin?

Broadcom expects to maintain a non-GAAP operating margin of 66% in the fourth quarter.

How much revenue growth is Broadcom still forecasting?

Overall fourth-quarter revenue is projected to grow 93% year over year despite missing the specific dollar estimate analysts had modeled.

What is driving concern about Broadcom’s custom chip business?

Reports point to intensifying competition in custom AI processors as the reason behind the softer-than-expected guidance.

Sources

  • The Globe and Mail — Broadcom forecasts quarterly revenue below estimates, signalling intense competition. theglobeandmail.com
  • Investing.com — Broadcom slips as soft Q4 guide overshadows strong growth outlook. investing.com

Related coverage on Tamara News: Nvidia’s Hugging Face acquisition and Anthropic’s Nvidia compute deal.

One Earnings Call Added Billions to GitLab’s Market Value

GitLab shares jumped nearly 20% this week. The software company beat Wall Street’s second-quarter estimates and raised its full-year guidance. The GitLab stock earnings surge came on the back of accelerating enterprise demand and AI-linked product growth, according to TIKR.

The numbers behind the GitLab stock earnings surge

Net revenue reached $286.3 million, up 21.3% year over year. That beat analyst expectations by roughly 475 basis points. Adjusted earnings per share came in at $0.24, well above the $0.18 consensus estimate. New annual recurring revenue grew more than 40% from a year earlier. GitLab’s net retention rate climbed to 117%. That means existing customers spent noticeably more than they had the year before.

GitLab stock earnings surge
Enterprise software earnings have become a key signal for where AI-linked spending is heading.

First-order bookings more than doubled year over year. That signals GitLab’s expanded sales organization is converting new prospects faster than before. It is not just renewing its existing base.

What drove the growth

GitLab pointed to two forces behind the results. It made targeted investments in its sales team. It also saw rising interest in AI-assisted development tools built into its platform, sometimes called Flex internally. Existing customers are adopting these newer offerings faster than GitLab’s older, standalone tools grew in past years. The company’s own retention and bookings figures support that claim.

Yahoo Finance noted the results landed at a tricky moment. Investors have grown skeptical of software valuations broadly this year. That makes a clean beat-and-raise quarter matter more for sentiment than it might have a year ago.

The guidance that moved the stock

GitLab raised its full fiscal year 2027 guidance. It now expects revenue of $1.129 billion to $1.133 billion. Adjusted earnings per share guidance rose to $0.85 to $0.87. Both figures sit modestly above what Wall Street had already priced in. Raising guidance carries more weight than beating one quarter alone. It signals management’s confidence extends beyond a single strong stretch.

What GitLab does with the momentum

GitLab’s next test is simple to state but hard to answer. Can the pace of new bookings and retention hold through the back half of its fiscal year? Software buyers often tighten budgets ahead of their own year-end planning. Investors will also watch how fast GitLab’s AI-linked features convert from early adoption into larger contracts. That conversion is what ultimately justifies the higher guidance the company just gave.

Rivals in the DevOps space will likely point to GitLab’s results as proof that AI-assisted coding tools are translating into real revenue, not just hype, which could accelerate competitive spending across the sector.

How this fits the software sector’s mood

Software stocks have had a rough year overall. Investors have punished companies that missed growth targets, even by small margins. That backdrop makes GitLab’s beat stand out more than it might have in a stronger market. A clean quarter, paired with raised guidance, tells investors that at least one enterprise software name is not losing momentum to AI-native competitors.

Rivals in the DevOps space will study GitLab’s numbers closely. Several have leaned on AI messaging without yet showing matching revenue growth. GitLab’s results give the market a concrete data point. AI-linked features can convert into real, measurable bookings, not just marketing language.

Fund managers who track software earnings say GitLab’s quarter will likely raise the bar for what counts as an acceptable result this season. Competitors reporting in the coming weeks will face comparisons to these numbers whether they want them or not.

A single strong quarter rarely erases a year of investor skepticism on its own. GitLab will need to repeat this kind of beat-and-raise performance at least once or twice more before the market fully credits the AI-driven growth story management is now telling.

The next earnings call, roughly three months from now, will carry more weight than usual. A second consecutive beat would confirm the trend. A miss, even a small one, would revive the doubts this week’s results just quieted.

For now, the market has given GitLab the benefit of the doubt. That is a meaningfully different position than the one it was in before this week’s report landed.

Employees and recruiters in the DevOps space often watch stock reactions like this one for signals about hiring and budget plans. A strong quarter tends to loosen both, at least for a company that just told Wall Street it expects the growth to continue.

Reader questions, answered

How much did GitLab stock rise after earnings?

GitLab shares surged close to 20% following its second-quarter results. Some intraday reports showed gains as high as 25%.

What was GitLab’s actual revenue for the quarter?

Net revenue came in at $286.3 million, a 21.3% increase from the same quarter a year earlier.

Did GitLab beat earnings expectations?

Yes. Adjusted earnings per share of $0.24 beat the $0.18 consensus estimate from analysts.

What is GitLab’s new full-year guidance?

GitLab raised its fiscal 2027 guidance to $1.129 billion to $1.133 billion in revenue, with adjusted EPS of $0.85 to $0.87.

What is driving GitLab’s growth right now?

Company disclosures point to expanded sales investment and rising adoption of AI-linked development tools among existing customers.

Sources

  • TIKR — GitLab Stock Surges 20% After Strong Q2 Earnings Beat. tikr.com
  • Yahoo Finance — GitLab’s Earnings Beat Just Gave Software Bulls a New Test. finance.yahoo.com

Related coverage on Tamara News: Palo Alto Networks’ acquisition spree and Anthropic’s compute deal with Nvidia.

Two Drones Triggered a New Round of Airstrikes in Lebanon

The Israeli military struck southern Lebanon again on Sunday, September 6. Hezbollah had launched drones at Israeli soldiers stationed in the border security zone. The Israel strikes southern Lebanon operation killed four people and wounded at least 20 others, according to Lebanon’s Health Ministry, cited by Haaretz.

What triggered the strikes

The Israeli military said Hezbollah launched two drones toward its soldiers inside the security zone. No Israeli soldiers were hurt in that attack. Israel called the drone launch “a blatant violation.” It responded with strikes across several areas, including the town of Arab Salim. Residents there had already received evacuation warnings.

Israel strikes southern Lebanon
Southern Lebanon has seen repeated rounds of strikes despite a standing ceasefire framework.

Israeli warplanes also hit areas around Upper Nabatieh and the outskirts of Kfar Roummane. RTÉ reported the details. The Israeli military said the targeted buildings had links to Hezbollah activity, though it gave few specifics.

How this fits the current ceasefire arrangement

Israel and Hezbollah have operated under a US-mediated ceasefire for months. Both sides have accused each other of violations on a rolling basis throughout that period. Sunday’s exchange stands out. A drone launch from Hezbollah was followed within hours by lethal Israeli airstrikes. That is a sharper pattern than the smaller, contained incidents that have defined most of the ceasefire period.

France 24 separately reported that Lebanese sources put the recent death toll from Israeli strikes in the south at 13. That figure spans a broader window than Sunday’s operation alone. It shows how fluid casualty counts remain as reports arrive from multiple locations at once.

What civilians in the affected areas face

Residents of Arab Salim and nearby villages got Israeli evacuation warnings before the strikes hit. The Israeli military has used this pattern repeatedly in southern Lebanon operations. Warnings reduce casualties, but they do not eliminate them. Some residents cannot leave quickly. Others are unwilling to abandon homes and property on short notice.

What could happen next

Watch whether Hezbollah responds with more drone or rocket fire. That would test whether the broader ceasefire framework survives this exchange or unravels further. US mediators brokered the original ceasefire, and they have a direct interest in preventing a full return to open conflict. Diplomatic pressure on both sides typically intensifies within days of an incident this serious.

Lebanese officials will also face pressure to show they can restrain Hezbollah’s actions near the border, a demand Israel has repeated after nearly every recent incident in the security zone.

The ceasefire’s rocky history

The current Israel-Hezbollah ceasefire has never been fully quiet. Both sides have logged complaints against the other almost every month since it took effect. Most incidents involved small-scale fire that drew limited responses. Sunday’s exchange broke that pattern. A drone launch drew an immediate, lethal air campaign rather than a contained reply.

Lebanese officials have said before that they lack full control over Hezbollah’s actions near the border. Israeli officials counter that Lebanon’s government bears responsibility regardless. That disagreement has never been resolved, and it resurfaces after nearly every incident like this one.

Southern Lebanon’s residents live with that unresolved dispute day to day. Evacuation warnings like Sunday’s have become a recurring feature of life near the security zone. Each new round of strikes adds pressure on families deciding whether returning home is worth the risk.

US mediators face a narrow path here. Push Israel too hard and risk looking weak on security guarantees it has offered in the past. Push too little and risk watching the ceasefire erode incident by incident until it is a ceasefire in name only.

Hezbollah, for its part, faces its own balancing act. It answers to a domestic audience that expects a response to Israeli strikes, while also operating under a Lebanese government that wants to avoid a full return to war. That tension shapes how measured or aggressive its next move is likely to be.

Neither side has framed Sunday’s exchange as a turning point so far. History in this conflict suggests that framing can shift quickly if either side reads the other’s next move as an escalation rather than a one-off response.

For residents in the border villages, the label diplomats choose matters less than whether the strikes stop. That distinction, between a policy debate in distant capitals and daily life under threat, is easy to lose in coverage of any single incident.

Common questions

What started the September 6 strikes in southern Lebanon?

Hezbollah launched two drones at Israeli soldiers in the border security zone. Israel responded with airstrikes across several areas of southern Lebanon.

How many people were killed or wounded?

Lebanon’s Health Ministry reported four people killed and at least 20 wounded in the strikes, according to Haaretz.

Is there still a ceasefire between Israel and Hezbollah?

A US-mediated ceasefire has held for months. Both sides have repeatedly accused each other of violations, and incidents like this one test whether it holds.

Did Israel warn residents before striking?

Yes. The Israeli military issued evacuation warnings for areas including Arab Salim ahead of the strikes.

Where exactly did the strikes hit?

Reported strike areas include Arab Salim, Upper Nabatieh and the outskirts of Kfar Roummane in southern Lebanon.

Sources

  • Haaretz — IDF says it struck southern Lebanon after Hezbollah launched explosive drones. haaretz.com
  • RTÉ — Israel launches fresh strikes on southern Lebanon. rte.ie

Related coverage on Tamara News: the Strait of Hormuz tanker attack and Russia’s latest strikes on Kyiv.