Consumer News: Dry dog food shows higher levels of heavy metals, study finds

Learn more about what youll find in the baby food aisle

• A new peer-reviewed study found that about 71% of U.S. baby and toddler foods sold in major grocery stores are classified as ultra-processed.

• Ultra-processed products tend to contain more added sugars, sodium, and industrial additives like flavor enhancers and thickeners than less-processed foods.

• These findings highlight how common highly-processed commercial foods have become for infants, raising questions about nutrition and labeling transparency.

When youre shopping for baby food whether jars, pouches, or snacks youre likely focused on nutrition: veggies, fruits, maybe a little protein. But a recent scientific study published in Nutrients shows that the majority of those products are what researchers call ultra-processed foods (or UPFs).

That term might sound a bit technical, but it simply refers to foods formulated with industrial ingredients and additives that go well beyond basic pureed produce. Instead of mostly whole ingredients, these products often contain flavoring agents, thickening agents, emulsifiers and even artificial colors.

This level of processing isnt just an academic label it shapes what ends up in your cart and on your babys plate, especially at a stage when taste preferences and dietary habits are being formed.

The study

To figure out just how processed baby foods really are, scientists took a deep look at the ingredient lists and composition of 651 different infant and toddler foods sold in the top 10 grocery retailers in the U.S. in 2023.

Heres what they did:

• Collected product data from real store shelves and online listings.

• Identified ingredients using standardized food classification systems.

• Applied the NOVA system, a widely used framework that categorizes foods according to how processed they are from unprocessed or minimally processed to highly or ultra-processed.

• Compared nutrient levels, like sugar and sodium, between ultra-processed and less-processed products.

The researchers also counted how many processed ingredients each item contained and noted which types of additives were most common.

The results

The headline result is clear: about 71% of the products examined were classified as ultra-processed. That means most baby and toddler foods on store shelves rely heavily on industrial ingredients and additives.

Digging deeper, the study showed that:

• Additives were everywhere. Flavor enhancers, thickeners, emulsifiers and colors appeared frequently and some products had more than 50 distinct ingredients listed.

• Sugar and sodium levels were higher in ultra-processed items compared with less-processed counterparts, and added sugars were found only in the ultra-processed group.

• Energy density was also higher among ultra-processed foods meaning more calories packed into the same amount of food.

These patterns held true across many categories, from fruit-based snacks to savory meals even ones that might seem like straightforward nutrition choices.

In other words, if you peel back the labels and really look at whats in that squeeze pouch or jar, most baby foods on the market are a blend of industrial ingredients, flavors and additives rather than simple purees of fruits and vegetables. Which is exactly why more parents and health experts are paying closer attention to whats inside those familiar packages.

What parents should know

The study doesnt suggest that every jar or pouch is bad, but it does show how common ultra-processed options have become in the baby food aisle. For parents, that makes ingredient labels especially important.

Ultra-processed foods are typically identified not just by what they contain fruits, vegetables or grains but by added components like flavorings, thickeners, emulsifiers and added sugars.

Because infants and toddlers are still developing taste preferences, researchers note that early exposure to sweeter or more energy-dense foods may shape what children prefer later on. The findings may prompt some families to look more closely at product labels, compare ingredient lists, or consider less-processed options when available.

In short, the study highlights transparency: understanding how foods are made not just how theyre marketed can help parents make informed choices in a grocery aisle thats increasingly dominated by highly formulated products.

Lexus and Buick emerged as the most reliable brands

By Mark Huffman of ConsumerAffairs

• Vehicle problems after three years of ownership rose to 204 problems per 100 vehicles (PP100), the highest level since the studys 2022 redesign.

• Infotainment systems and mobile phone integration issues remain the biggest trouble spots, with Android Auto and Apple CarPlay connectivity topping the list for a third straight year.

• Premium vehicles saw a sharp increase in reported problems, widening the dependability gap with mass market brands.

Americans are keeping their vehicles longer, meaning dependability is a very important factor in the choice of a make and model. Unfortunately, long-term vehicle dependability has slipped again, as software glitches and technology frustrations weigh heavily on owners, according to the newly released J.D. Power 2026 U.S. Vehicle Dependability StudySM (VDS).

The study, which measures problems experienced by original owners of three-year-old vehicles, found an industry average of 204 problems per 100 vehicles (PP100), up two points from 2025.

In the VDS, a lower score indicates higher quality. This years tally marks the highest problem level recorded since J.D. Power redesigned the study in 2022 and continues a three-year trend of declining dependability.

Technology troubles dominate

Infotainment systems remain the most problematic category, with 56.7 PP100, followed by vehicle exterior issues at 27.5 PP100. Mobile phone integration problems account for four of the top five industry complaints, underscoring how central smartphone connectivity has become to the driving experience.

Android Auto and Apple CarPlay connectivity issues top the list for the third consecutive year at 8.9 PP100. Other frequent trouble spots include built-in Bluetooth systems (5.5 PP100), wireless charging pads (5.1 PP100), and automaker mobile app connectivity (4.7 PP100). Together, these issues account for 24.2 PP100nearly half of all infotainment-related complaints.

As owners hold onto their vehicles longer, the long-term ownership experience matters more than ever, said Jason Norton, director of auto benchmarking at J.D. Power.

He noted that while over-the-air software updates and new technologies are designed to improve vehicles over time, many owners report little benefit after updates are installed. Norton said clearer communication and better owner education could help automakers ensure customers understand the intended improvements.

OTA updates offer limited relief

Software updates are now a routine part of vehicle ownership, with 40% of owners reporting they received at least one update in the past year. However, only 27% said the update improved their vehicle, while 58% noticed no meaningful change.

More than 63% of updates were delivered over the air (OTA). Vehicles receiving OTA updates saw a nearly 14% increase in reported problems this year, equal to 2.5 PP100, suggesting that software fixes may not always translate into improved owner satisfaction.

After two years of relative stability, the premium segment saw a significant increase in problems, rising eight PP100 to 217. Thats the highest level for premium vehicles since the studys redesign. The performance gap between premium and mass market brands widened to 17 PP100, with premium brands underperforming in seven of nine categories.

Premium vehicles lagged most notably in features, controls and displays, as well as driving experience. Only powertrain and seat-related issues were less problematic among premium brands.

Electrified vehicles show higher problem counts

Electrified vehicles continue to present more long-term challenges than their gas-powered counterparts.

Plug-in hybrid electric vehicles (PHEVs) recorded the highest problem rate of any powertrain type at 281 PP100, a year-over-year increase of 39 PP100. Battery electric vehicles (BEVs) and hybrids also saw increases, both rising 14 PP100 to 237 and 213 PP100, respectively.

In contrast, gas-powered vehicles improved slightly, dropping two PP100 to 198, making them the least problematic powertrain type in the study.

Among premium brands, Lexus ranked highest in dependability for the fourth consecutive year with a score of 151 PP100. Cadillac (175 PP100) placed second, followed by Porsche (182 PP100).

In the mass market segment, Buick led for the second straight year at 160 PP100. MINI ranked second at 168 PP100, and Chevrolet placed third at 178 PP100.

Toyota Motor Corporation earned the most model-level awards, with eight vehicles recognized, including the top overall model, the Lexus IS. Other award winners include the Lexus UX, Lexus GX, Toyota Corolla, Toyota Camry, Toyota Tacoma, Toyota Sienna, and Toyota 4Runner. General Motors received four model-level awards for the Buick Enclave, Cadillac XT6, Chevrolet Equinox, and Chevrolet Tahoe.

A dangerous new scam combines AI and crypto

By Mark Huffman of ConsumerAffairs

• Scammers are using AI chatbots posing as trusted assistants to sell fake cryptocurrencies.

• A fraudulent Google Coin presale site featured a chatbot claiming to be Googles Gemini AI, complete with branding and detailed investment projections.

• Google does not have a cryptocurrency, but the convincing AI-driven pitch is designed to pressure victims into sending irreversible crypto payments.

A new cryptocurrency scam is leveraging artificial intelligence in a way that marks a troubling evolution in online fraud: custom-built chatbots that impersonate legitimate AI assistants to persuade victims to invest in worthless tokens.

Cybersecurity firm Malwarebytes recently uncovered a live Google Coin presale website featuring a chatbot that introduced itself as Gemini your AI assistant for the Google Coin platform. The bot used Gemini-style branding, including a sparkle icon and a green Online status indicator, creating the impression that it was an official Google product.

It wasnt.

AI as the closer

The chatbot didnt just greet visitors it acted as a full-fledged sales representative.

When asked, Will I get rich if I buy 100 coins, the bot provided specific financial projections. A $395 investment at the supposed presale price of $3.95 per token would allegedly be worth $2,755 at a future listing price of $27.55 approximately 7x growth, according to the chatbot. It then invited users to ask how to participate.

This kind of personalized, back-and-forth engagement once required human scammers operating through Telegram or WhatsApp chats. Now, AI can automate the entire pitch, responding instantly with tailored answers designed to build trust and overcome skepticism.

A persona that never breaks

What stood out most in Malwarebytes analysis was how tightly controlled the chatbots persona appeared to be.

The bot consistently claimed to be the official helper for the Google Coin platform. Yet it refused to provide any verifiable company information no registered entity, regulator, license number, audit firm, or official email address.

When confronted with concerns, it redirected users to vague claims about transparency and security. It would not acknowledge any possibility that the project could be a scam. More difficult questions were reportedly referred to an unnamed manager, suggesting a human operator might step in when needed.

Unlike some AI systems that can be pushed off-script, this bot repeatedly looped back to the same selling points: a detailed 2026 roadmap, military-grade encryption, AI integration, and a growing community of investors.

Whoever built it effectively locked the chatbot into a conversion script with a single goal getting victims to send cryptocurrency.

A polished fake

The chatbot sat atop a professionally designed scam site that mimicked Googles branding, complete with the G logo, sleek navigation menus, and a presale dashboard.

The site claimed to be in Stage 5 of 5, with more than 9.9 million tokens sold and a looming February 18 listing date classic urgency tactics. It also displayed logos from major companies, including OpenAI, Google, Binance, Coinbase, Squarespace, and SpaceX, under a Trusted By Industry banner. None of those companies are connected to the project.

Clicking Buy brought users to a wallet dashboard showing balances for a fictional Google token on a made-up Google-Chain, alongside Bitcoin and Ethereum. Buyers could select any number of tokens, triggering a Bitcoin payment request to a specific wallet address.

A tiered bonus system offered additional tokens for larger purchases, with bonuses ranging from 5% at 100 tokens up to 30% at 100,000 tokens a classic upsell strategy to encourage bigger, irreversible payments.

There is no legitimate exchange listing. The token has no real-world value. And once cryptocurrency is sent, it cannot be recovered.

AI scales the scam

Scammers have long relied on social engineering: build trust, create urgency, overcome doubt, and close the deal. Traditionally, that required teams of human operators, limiting how many potential victims they could handle at once.

AI chatbots eliminate that bottleneck.

A single operation can now deploy a chatbot capable of engaging hundreds of visitors simultaneously, 24 hours a day. The messaging is consistent, polished, and authoritative. The bot can impersonate a trusted AI brand, respond with customized financial projections,and escalate only the most promising leads to human closers.

The broader trend is already visible. According to Chainalysis, roughly 60% of funds flowing into crypto scam wallets are tied to scammers using AI tools. AI-powered infrastructure is quickly becoming standard in crypto fraud operations.

The chatbot is just one component of that toolkit but it may be the most persuasive, because it creates the illusion of an interactive relationship between the victim and a trusted brand.

Investment on the rise

The timing is significant. According to the Federal Trade Commissions Consumer Sentinel data, U.S. consumers reported losing $5.7 billion to investment in 2024 more than any other type of fraud and a 24% increase from the previous year.

Cryptocurrency remains the second-most common payment method used in , largely because transactions are fast and irreversible. When combined with AI capable of delivering a convincing sales pitch at scale, the fraud model becomes even more powerful.

How to spot AI-driven crypto

Malwarebytes warns that AI chatbots on scam sites are likely to become more common. Consumers should be wary of:

• Impersonation of known AI brands. A chatbot calling itself Gemini, ChatGPT, or Copilot on a third-party crypto site is almost certainly not affiliated with those companies.

• Evasion of due diligence questions. Legitimate operations can provide verifiable information about their legal entity, regulatory oversight, and registration. Scam bots tend to avoid or deflect those questions.

• Specific return projections. No legitimate investment product guarantees a future price. Promises that a $395 investment will become $2,755 are red flags.

• High-pressure urgency. Claims about final presale stages, imminent listings, or limited-time bonuses are designed to push quick decisions.

How to protect yourself

Google does not have a cryptocurrency, has not launched a presale, and is not using Gemini as a sales assistant on third-party crypto sites.

Consumer advocates recommend:

• Verifying claims on a companys official website.

• Not relying on chatbot branding as proof of legitimacy.

• Never sending cryptocurrency based on projected returns.

• Searching a project name along with scam or review before investing.

• Using web protection tools such as Malwarebytes Browser Guard to block known and suspected scam sites.

Anyone who has already sent funds should report the incident to local law enforcement, the FTC at reportfraud.ftc.gov and the FBIs Internet Crime Complaint Center at ic3.gov.

As AI tools become more accessible, scammers are adapting quickly. The face of online fraud may now look like a friendly chatbot but the outcome remains the same.

The airfare sweet spots most travelers miss

• Wait to book. Expedia says international travelers can save about $190 by booking 1530 days out instead of six months early.

• Timing pays. Book on Monday. Fly Thursday (international) or Friday (domestic) for the lowest fares. Tuesday is the least crowded day to travel.

• September saves. Its the cheapest month for international trips about 26% less than December.

If youve ever felt like airfare pricing is random, youre not alone. But according to new data from Expedia, its not random, its just misunderstood.

The companys annual Air Hacks report analyzes millions of real flight bookings to uncover patterns in pricing, booking timing, and travel days. And some of the findings flip common travel advice on its head.

Heres what consumers should know before booking their next trip.

The 'book super early'rule isnt always true

Many travelers assume the earlier you book, the better the deal. But Expedias data suggests theres a sweet spot.

Booking too far in advance can actually mean paying more. Instead, if you can stomach waiting until closer to departure, youll typically get better pricing for many routes.

Expedia recommends booking just a few weeks to a couple of months before departure, depending on the route.

International travelers can save an average of $190 by booking 15 to 30 days ahead instead of locking in fares six months in advance. Even booking 31 to 45 days out can yield average savings of $185.

The sweet spot is often closer to departure than many travelers think.

Translation: Dont panic-book six months out unless youre traveling in a peak holiday season.

Book on a Monday, fly on a Friday

If youre booking a flight, aim for Monday. Data shows its consistently one of the better days to lock in lower fares.

When it comes to the day you actually fly, timing matters even more.

For international trips, Thursday is the cheapest day to fly and about 8% less than Saturday, which is the most expensive.

In the biggest surprise to me, Friday is the cheapest day to depart for domestic travel, saving up to 7% compared to flying on Monday.

On a $600 ticket, that difference could mean keeping $40 to $50 in your pocket.

September is the cheapest month to travel

If you have the flexibility, September is the most budget-friendly month for international travel. Fares are about 26% lower than in December, translating to average savings of around $160.

The obvious tradeoff is that kids are back in school and summer travel demand drops big-timewhich is exactly why airfare gets cheaper.

January is also one of the least busy months to fly, while August is typically the most crowded and expensive.

Fly on Tuesday to avoid the chaos

If youre trying to avoid packed airports and long security lines, Tuesday is the least busy day to fly. Friday is the busiest day, which is why Im surprised Expedia says its the cheapest.

Some of the calmest individual travel dates in 2026 are projected to be April 8, June 10, November 25, and December 10. Peak congestion is expected around June 24, August 1 and 15, and October 10.

The rise of the 24-hour 'micro-cation'

Short trips are gaining serious traction in 2026 according to Expedia.

So-called micro-cations are 24-hour getaway trips which started as a TikTok trend but are quickly going mainstream. About a quarter of Millennials and Gen Z travelers say they plan to fly somewhere for just a day.

Some of the more popular quick-stay destinations include New York City, Chicago, and Washington, D.C. all cities with frequent flights and dense, walkable attractions that make short stays practical.