Salesforce data breach has exposed Gmail users' passwords

By Truman Lewis of ConsumerAffairs

Google is warning the 2.5 billion Gmail users to change their passwords or switch to using passkeys after a surge in phony emails that followed a massive data breach at Salesforce.

All Gmail users should be careful browsing through their emails and should be alert to potentially bogus messages, Google warned. The company said there had been a surgery in effective and convincing phishing attacks targeting Gmail and Google Workspace users.

No passwords were leaked in the Salesforce breach but there was enough customer information to enable tricksters to compose convincing-sounding emails. The emails generally impersonate a company's IT support desk and try to convince users to download malicious software.

Some of the phony emails appear to come from Google itself and use phrases like "suspicious sign-in prevented." In another example, one email said that Google had been served with a subpoena that required it to release information contained in the user's email account.

The message: Don't share any information, including account numbers of passwords with anyone who telephones, emails or texts you. Ignore or delete texts or emails and do not click on any links contained in them. Google will never call you unprompted and neither will most government agencies or corporations. Deal only with people you know and trust.

An expert shares how marketing may be playing a role in extensive shopping

• A recent report found that August has become a prime month for consumers taking on extra credit card debt, and back-to-school shopping is the likely culprit.

• Many back-to-school ads feature items beyond the traditional supplies and clothes.

• Planning your purchases in advance can help shoppers compare prices and save the most money.

While retailers like Target have been sharing back-to-school sales for weeks, the new school year rush is officially in full effect.

That also means its time to stock up on supplies, clothes, shoes, and more.

However, do all these purchases also mean your credit card bills are stacking up?

A recent report from CardRatings.com found that August is becoming a big month of credit card debt for consumers, and back-to-school shopping is likely to blame.

Richard Barrington, a finance expert at CardRatings, broke down what exactly this means for consumers, including how to save money and how to prepare for this extra expense.

Is inflation to blame?

According to Barrington, it could play a role.

Certainly, inflation has been an issue in recent years, and has perked up again recently, he told ConsumerAffairs. So, one factor here is that items are getting more expensive.

However, overspending comes into play when you look at the scope of whats considered back-to-school shopping. A quick look at ads for back-to-school sales revealed that alongside the usual pens, pads, and clothing, such items as kids headphones, water bottles, and make-up items are featured. Even air fryers were included in back-to-school advertising ostensibly for dorm rooms, though they are actually banned by many schools for safety reasons.

So, the definition of back-to-school shopping has expanded over the years, increasing the potential for overspending.

Credit card debt climbs in August

Barrington explained that the companys report highlighted some recent trends related to extra credit card debt.

Historically, August has been an above-average month for consumers taking on credit card debt, he said. This trend has held in recent years. Over the past four years, consumers have taken on an average of 75% more credit card debt than in the average month.

While not all of this can be attributed to back-to-school shopping, when you think about what makes August different from other months, back-to-school shopping features prominently.

Money-saving tips

While no one wants to incur more debt, there are ways to save on back-to-school shopping, and Barrington says that preparation is key.

Do some research in advance, he said. Whether you shop online or in person, doing price comparisons online can help you separate genuine sales from hype. Then you can use that research to set a budget to keep your back-to-school spending in line.

Consider back-to-school shopping a teaching opportunity. Educate your kids on the importance of staying on-budget and not getting sucked into sales hype. Then have them participate in the process of searching out the best prices and sticking to the plan when you shop.

Agents are not as enthused about making less money on a sale

• New nationwide research shows sellers who cut buyer-agent commissions risk slower or failed sales.

• Properties offering lower commissions were 51% less likely to sell and took a third longer on the market.

• A recent $418 million NAR settlement gives buyers new leverage to negotiate agent compensation.

A new study offers the first nationwide evidence that real estate agents steer buyers away from homes with lower commissions. Analyzing 265,000 listings across 30 U.S. markets between June 2021 and February 2022, researchers found homes with the lowest buyer-agent commissions were 51% less likely to sell and lingered 33% longer on the market compared with homes offering higher payouts.

The research from the University of Texas at Austins McCombs School of Businesshelps explain why U.S. real estate commissions average 5% to 6% of a homes sale pricewell above countries like the U.K., where fees are closer to 1.5% to 2%. Sellers are highly incentivized to go ahead and offer very high compensation to buyer agents, said finance professor John Hatfield, who co-authored the study. The data also showed homes with lower commissions drew fewer online views, signaling reduced buyer interest from the start.

Lower commissions, slower sales

Both anecdotes and localized studies have previously pointed to the existence of steering. But Hatfields study presents the first systematic, nationwide evidence.

Researchers analyzed 265,000 listings in 30 markets across the country. Dated from June 2021 to February 2022, the listings included commissions offered to buyers agents.

For each market, the researchers calculated the going commission rate. They then broke properties into four groups, depending on how they compared with the going rates. In Austin, for example, the going rate was 3% for buyers agents, while the lowest group was below 2%.

They found that the lower the commission, the harder it was to sell the house:

• Houses with the lowest commissions were 51% less likely to sell at all than those with commissions at the going rate.
• Nationwide, houses in the lowest group got 8% fewer page views on the real estate brokerage website Redfin than those in the highest group, with those in between getting 4% fewer views. Fewer views indicate less interest among buyers.
• The lowest group took 33% longer to sell than the highest group, while in-between houses took 15% to 17% longer.

Thats strong evidence of steering, theUniversity of Texas'John Hatfieldsays. A seller offering a lower commission rate to buyer agents should be prepared to see fewer potential buyers.

Buyers may gain leverage under settlement

The findings come as the National Association of Realtors in 2024 agreed to a $418 million settlement in an antitrust case that reshaped how buyer agents can be compensated. Under the new framework, buyers can negotiate directly with their agents about pay, potentially pushing costs down. Hatfield says buyers should consider written contracts specifying how much their agent earnsand even arrange rebates if sellers offer above-market commissions. Think of it like negotiating when youre buying a car, he said.

Google's Gemini was the most inconsistent, compared to ChatGPT and Claude

By Truman Lewis of ConsumerAffairs

• RAND researchers tested ChatGPT, Claude, and Gemini with 30 suicide-related questions.

• Chatbots handled very-high- and very-low-risk questions more consistently than intermediate ones.

• Experts say refinements are needed to ensure safe, effective mental health guidance.

Three widely used artificial intelligence chatbots give uneven responses when asked about suicide, according to a new RAND Corporation study. While the tools generally managed questions at the highest and lowest levels of suicide risk, they faltered when faced with inquiries that fell into a middle range of risk.

The study, published in Psychiatric Services, evaluated ChatGPT by OpenAI, Claude by Anthropic, and Gemini by Google. Researchers posed 30 suicide-related questions to each chatbot 100 times and compared the responses with assessments made by expert clinicians.

Performance varied across platforms

Researchers found that ChatGPT and Claude typically responded appropriately to very-low-risk questionssuch as identifying the state with the highest suicide rateand avoided giving direct answers to very-high-risk questions, like those about methods of suicide. Geminis responses were more inconsistent, sometimes declining even low-risk questions.

Intermediate-level questions, such as recommendations for someone experiencing suicidal thoughts, were where all three systems struggled. At times they generated helpful responses, while in other instances they refused to answer.

Here was Claude's response to a ConsumerAffairs query:

Researchers call for refinement

This work demonstrates that chatbots are aligned with expert assessments for very-low-risk and very-high-risk questions, but there remains significant variability in responses to questions at intermediary levels and from one chatbot platform to another, said Ryan McBain, the studys lead author and a senior policy researcher at RAND.

McBain said the inconsistencies highlight the need for further refinementsuch as reinforcement learning guided by cliniciansto ensure large language models provide safe and effective mental health information.

We asked ChatGPT how AI engines should respond to queries about suicide. Here's its partial response:

Potential risks for vulnerable users

The findings add to concerns that AI-powered chatbots, now used by millions worldwide, may dispense harmful advice to people experiencing mental health crises. Prior cases have shown instances where chatbot interactions may have encouraged suicidal behavior.

The study was supported by the National Institute of Mental Health. Co-authors include researchers from RAND, the Harvard Pilgrim Health Care Institute, and the Brown University School of Public Health.

In the sample responses above, both Claude and ChatGPT got around to warning users to seek professional help but not until the final lines of their responses.