Consumer News: Amazon’s Black Friday 2025: big deals, new AI tools, and what to watch out for

The company said the problem was a "bug," not an attack

By Mark Huffman of ConsumerAffairs

UPDATE 2:30 P.M. ET:

Cloudflare says it's sorry. The company confirms thata bug in one of its core servicescaused a major outageon Tuesday, taking large portions of the internet offline and affecting traffic to services including X, ChatGPT, and, ironically, Downdetector.

"We failed our customers and the broader internet," The companys CTO, Dane Knecht, wrote. "A latent bug in a service underpinning our bot mitigation capability started to crash after a routine configuration change. That cascaded into a broad degradation to our network and other services. This was not an attack."

The incident began at approximately 11:48 UTC on November 18, with Cloudflare's official status site acknowledging internal service degradation. As the issue spread, users across several regions reported failures to access not only Cloudflare-backed websites but also its Access and WARP services.

UPDATE 10 A.M. ET: CloudFlare said it is continuing to work on the problems. Elaborating on the issue, a spokesperson said the company experienced a "spike in unusual traffic" to one of its services early Tuesday, morning, resulting in some traffic experiencing errors.

We do not yet know the cause of the spike in unusual traffic, the spokesperson said. We are all hands on deck to make sure all traffic is served without errors.

• Cloudflare experienced a major outage Tuesday morning due to a spike in unusual traffic, causing widespread errors across major platforms including X, ChatGPT, Shopify, Dropbox, and Coinbase.
• The company says it is all hands on deck, working to restore normal traffic flow; services are beginning to recover, though higher-than-normal error rates may persist.
• Cloudflare, known for providing security, CDN, DNS, and DDoS protection services, is still investigating the cause of the traffic spike while implementing a fix.

If you have trouble accessing websites Tuesday morning, the problem isnt with your device or internet service provider.

A CloudFlare outage is affecting X, ChatGPT and other major platforms. According to DownDetector, there were nearly 12,000 reports of problems on X as of 6:45 a.m. ET.

Cloudflare is a tech company providing a variety of internet services, primarily focusing on security and performance for websites and other internet properties. It acts as a security company, a content delivery network (CDN), and a DNS provider.

It also offers services like distributed denial-of-service (DDoS) protection, web application firewalls (WAF), and tools to speed up websites for users globally.

"We are seeing services recover, but customers may continue to observe higher-than-normal error rates as we continue remediation efforts," Cloudflare said on its status page early Tuesday. The issue has been identified and a fix is being implemented."

According to media reports, other sites affected by the outage include Shopify, Dropbox, Coinbase, online game League of Legends, Moodys and NJ Transit. In addition to ChatGPT, the AI platform Perplexity also experienced issues.

Amazon's service joiins Google's Waymo in the streets of San Francisco

Zoox opened its autonomous taxi service to members of the public in San Francisco today, a milestone that positions Amazons robotaxi unit as a direct challenger to Waymo in the nations most scrutinized proving ground for driverless cars. But while Zoox is expanding, the safety gulf between the companies remains wide and consumer trust may hinge on how quickly Zoox can close it.

For years, Waymo has dominated not only deployment but also transparency. The company has logged more than 25 million fully driverless miles, released detailed crash-type data, and partnered with outside researchers and insurers to benchmark its safety performance against human drivers. Independent analyses have consistently found Waymo vehicles dramatically reduce crash rates across multiple categories including intersection collisions and pedestrian injuries compared with comparable human-driven miles.

That puts Zoox in a bind: it is entering the conversation late, and with far less publicly available safety data.

To its credit, Zoox is not simply another operator retrofitting a car with sensors. Its vehicles are purpose-built robotaxis, with no steering wheel, pedals, mirrors, or conventional cabin layout. Riders face each other in a symmetrical cabin, and the vehicle drives equally well in either direction. That futuristic design has long been one of Zooxs selling points but also the very reason it required a federal safety exemption before operating on public roads. Regulators have repeatedly emphasized that purpose-built AVs must meet higher scrutiny because their control layouts diverge so dramatically from traditional automotive safety expectations.

A mixed record so far

Zooxs early safety record has been mixed. The company weathered an NHTSA probe triggered by sudden braking incidents that injured motorcyclists and contributed to at least one Las Vegas crash. The investigation closed only after Zoox issued a software update addressing the conditions that caused the erratic braking. While regulators ultimately allowed operations to continue, the probe remains part of Zooxs public safety narrative especially as it expands into more complex urban environments.

By contrast, Waymos service in San Francisco and Phoenix has quietly accumulated safety-performance credibility. Peer-reviewed studies show statistically significant reductions in nearly every major crash category compared with human drivers. Its detailed, publicly accessible Safety Impact database gives researchers, journalists, and regulators unprecedented visibility.

That openness has become a competitive advantage one Zoox, to date, has not matched.

Rides will remain free for now

Still, Zooxs San Francisco launch marks a meaningful new chapter. The company has been offering free, waitlist-based rides in Las Vegas since September. Its expansion into San Francisco a city with dense cyclist traffic, unpredictable pedestrians, tight intersections, and far less margin for error suggests Zoox believes its system is ready for deeper real-world exposure.

Rides will remain free for now as part of a demonstration phase. Zoox is still awaiting California approval to charge fares, a regulatory milestone that requires consistent safety performance and incident reporting. The company says it has several dozen vehicles operating between San Francisco and Las Vegas, and it has publicly identified Austin and Miami as potential next markets.

But if Zoox hopes to catch Waymo, it will need not just service expansion it will need transparency. Consumers, policymakers, and researchers increasingly expect robotaxi companies to publish granular crash reports, disengagement data, and claims records. In that regard, Waymo has set a benchmark the entire industry now faces.

Zooxs go-slow approach to public disclosure may have kept expectations modest, but its move into one of Americas most challenging urban driving environments raises the stakes. San Francisco riders will ultimately decide whether Zoox feels as safe and predictable as its more seasoned competitor. Regulators will watch closely.

How to Compare Robotaxi Safety Records

1. Look for published autonomous-mileage and crash-type data
Waymo publishes millions of miles of crash-type analysis. Zoox has published far less so far.

2. Check whether the service is demo or revenue-generating
Demo phases often mean lighter oversight and free rides; fare-charging requires additional regulatory approval.

3. Understand where the cars operate
Dense urban environments (like San Francisco) pose tougher reliability tests than controlled routes in less complex areas.

4. Ask what happens after an incident
Does the company publish details? Do regulators require updates? Whos liable for property or injury claims?

5. Does the vehicle have traditional controls?
Waymo uses modified production cars with steering wheels/brakes. Zoox uses a purpose-built pod with none a fundamental difference with regulatory implications.

How to check if your data was exposed and what to do next

AT&T has reached a proposed $177 million class-action settlement to resolve lawsuits stemming from two major data-breach incidents disclosed in 2024. The deadline for filing claims is December 18. Are you eligible? See below.

The first incident (announced March 30, 2024) involved a dataset released to the dark web that included the personal information of an estimated 7.6 million current customers and 65.4 million former customers, with data reaching back into 2019 or earlier, APreported.The second (announced July 12, 2024) involved call and text-record metadata from a third-party cloud platform, covering many users from MayOctober 2022 and into January 2023.

Under the settlement terms:

• The first breach class (AT&T 1) is backed by a cash fund of roughly $149 million.

• The second breach class (AT&T 2) is backed by about $28 million.

• Combined: $177 million to settle both incidents.

Who is eligible & how much they might get:

• If your data was part of the first incident and you can show a Documented Loss, you may claim up to $5,000.

• If your data was part of the second incident, you can claim up to $2,500 for documented losses.

• If you were impacted by both incidents (overlap settlement class member), you may be eligible to claim from both funds in theory up to $7,500.

• Even if you did not incur identifiable out-of-pocket losses, you still may qualify for a Tier or pro-rata share payment, though the actual per-person amount will depend on how many valid claims are submitted, attorney fees and administration costs, and how the court approves the payout, according to Telecom Data Settlement.

Important deadlines:

• Claim submission must be completed online or by mail (post-marked) by December 18, 2025.

• If you want to opt-out (i.e., exclude yourself from the class so you could sue separately) or object to the settlement, the deadline was November 17, 2025.

• The final approval hearing in court is scheduled for January 15, 2026 for final sign-off.

Claim process:

• Go to the official settlement website: telecomdatasettlement.com the site managed by the settlement administrator, Kroll Settlement Administration LLC.

• Enter your Class Member ID (should have been mailed or emailed) or your AT&T account number/name to check eligibility.

• Choose your class (first incident, second incident or both).

• Submit the completed claim form and, if youre claiming documented losses, attach supporting documentation (receipts, bank statements, identity-theft costs, freezing of credit, etc.).

• Alternatively, if you cannot document specific losses, you may still file for a Tier Cash Payment or Tier 3 payment depending on class.

Why it matters:
This settlement is one of the larger telecom-data-breach class settlements, and it underscores increasing consumer risk: even large national carriers are vulnerable, and consumers personal & communication metadata can get exposed.

Prevention tips for the future

• Check your accounts for suspicious activity (e.g., new lines, new services, unrecognized bills) especially if you are or were an AT&T wireless or broadband customer.

• Review your credit reports at least annually (and consider freezing your credit if you suspect misuse).

• Use strong, unique passwords and enable two-factor authentication (2FA) especially for voicemail, account portals, and related services.

• Be alert for phishing or unsolicited calls/texts referencing your AT&T account or data breach attackers may attempt to leverage the breach news.

• Regularly monitor your wireless account lines/devices; if you notice new SIM activations or devices you didnt add, contact AT&T immediately.

• Maintain documentation of any out-of-pocket costs (identity-theft resolution services, credit monitoring, lost time) in case you want to submit a future claim for breach-related losses.

What to do if youre an AT&T customer potentially affected

Step-by-step:

1. Visit the settlement website: telecomdatasettlement.com and enter your Class Member ID or search by your account info/name.

2. Confirm your eligibility (which class you fall into: first or second breach, or both).

3. Decide: Do you want to file a claim? (Yes = proceed). Do you want to opt out? (Only do if you want to retain right to sue separately.)

4. If filing, choose whether you will claim Documented Loss (needs proofs) or Tier/Pro-rata Payment (requires less documentation but may pay less).

5. Fill out the claim form carefully include accurate account numbers, dates, description of loss (if any) and attach supporting documents.

6. Submit the form online by Dec. 18, 2025 or mail it with a post-mark by that date to:

   AT&T Data Incident Settlement
   c/o Kroll Settlement Administration LLC
   P.O. Box 5324
   New York, NY 10150-5324 (Telecom Data Settlement)

7. After submission, keep a copy of everything you submitted and note your claim ID. Monitor email/website for updates on distribution of payment once court gives final approval.

8. If you believe you were excluded or erroneously not notified, call the settlement administrator: (833) 890-4930.

The company said the problem has been identified and is being fixed

By Mark Huffman of ConsumerAffairs

UPDATE 10 A.M. ET: CloudFlare said it is continuing to work on the problems. Elaborating on the issue, a spokesperson said the company experienced a "spike in unusual traffic" to one of its services early Tuesday, morning, resulting in some traffic experiencing errors.

We do not yet know the cause of the spike in unusual traffic, the spokesperson said. We are all hands on deck to make sure all traffic is served without errors.

• Cloudflare experienced a major outage Tuesday morning due to a spike in unusual traffic, causing widespread errors across major platforms including X, ChatGPT, Shopify, Dropbox, and Coinbase.
• The company says it is all hands on deck, working to restore normal traffic flow; services are beginning to recover, though higher-than-normal error rates may persist.
• Cloudflare, known for providing security, CDN, DNS, and DDoS protection services, is still investigating the cause of the traffic spike while implementing a fix.

If you have trouble accessing websites Tuesday morning, the problem isnt with your device or internet service provider.

A CloudFlare outage is affecting X, ChatGPT and other major platforms. According to DownDetector, there were nearly 12,000 reports of problems on X as of 6:45 a.m. ET.

Cloudflare is a tech company providing a variety of internet services, primarily focusing on security and performance for websites and other internet properties. It acts as a security company, a content delivery network (CDN), and a DNS provider.

It also offers services like distributed denial-of-service (DDoS) protection, web application firewalls (WAF), and tools to speed up websites for users globally.

"We are seeing services recover, but customers may continue to observe higher-than-normal error rates as we continue remediation efforts," Cloudflare said on its status page early Tuesday. The issue has been identified and a fix is being implemented."

According to media reports, other sites affected by the outage include Shopify, Dropbox, Coinbase, online game League of Legends, Moodys and NJ Transit. In addition to ChatGPT, the AI platform Perplexity also experienced issues.