Category: Technology

Nvidia Investor Dilemma: Determining the Right Proportion for Your Portfolio

Adler MorisLeave a comment

Investor Gains and Growing Concerns

Outsize investments in Nvidia, the leading chipmaker driving the artificial intelligence revolution, have significantly boosted portfolio managers’ returns this year. However, these substantial bets pose increased risks should Nvidia’s soaring stock experience a downturn.

Since the beginning of 2023, Nvidia shares have surged approximately 785%, with a remarkable 160% rise this year alone. This growth has been driven by unprecedented demand for its AI chips, briefly making Nvidia the world’s most valuable company in June before Microsoft reclaimed the title.

Shift in Investment Strategies

The substantial rise in Nvidia’s stock price has led to a notable increase in the number of actively managed funds holding significant positions in the company. Morningstar data reveals that by the end of the first quarter, 355 actively managed funds held Nvidia stocks accounting for 5% or more of their assets, up from 108 funds in the same period last year. This trend reflects a strategy to either maximize profits or align with index benchmarks.

Jack Shannon, a senior analyst at Morningstar, notes, “There’s a mindset among some portfolio managers that they missed the boat on Apple or Microsoft and don’t want to be wrong on AI. They don’t want to sell.”

Market Concentration and Its Risks

Nvidia’s dominance is emblematic of a broader trend where a few major growth stocks have driven market gains. The chipmaker alone has contributed about a third of the S&P 500’s 17% gain this year, according to S&P Dow Jones Indices. The market has become increasingly concentrated, with only 24% of S&P 500 stocks outperforming the index in the first half of the year, as reported by BofA Global Research.

Funds holding Nvidia have benefited, with actively managed U.S. equity funds featuring the stock up an average of 16.3% in the first half of 2024, compared to 5.7% for those without Nvidia, Morningstar data shows.

Potential Pitfalls and Market Sensitivity

Despite these gains, holding a large portion of a portfolio in one stock can amplify risks. Nvidia’s current price-to-earnings ratio stands at 39.3 times forward earnings—50% higher than its industry median. Analysts express concerns about rising competition, supply-demand imbalances as Nvidia scales up production, and the stock’s high valuation, which could signal a potential downturn.

Phil Orlando, chief equity market strategist at Federated Hermes, warns, “Does having 6% or more of your portfolio in one stock create outsized risks? The answer is obviously, yes. The fact that one stock did take off like a rocket ship doesn’t mean it was smart to have that many eggs in one basket.”

Recent Market Movements and Investor Sentiments

Last week’s sharp rotation out of Big Tech stocks, triggered by cooler inflation data, highlighted the risks of concentrated positions. Nvidia’s shares fell nearly 6% in one day, marking its largest drop in over two weeks, while the Nasdaq 100 experienced a 2.2% decline. Both indices partially recovered the following day.

Technology-sector funds have shown the highest allocations to Nvidia, with several Fidelity funds holding over 18% of their assets in the stock. More diversified funds, like the Baron Fifth Avenue Growth fund and Fidelity Blue Chip Growth fund, also maintain significant positions in Nvidia.

Reflections and Regrets

Anthony Zackery, portfolio manager at Zevenbergen Capital Investments, has maintained a core position in Nvidia since 2016 but has occasionally trimmed it to manage risk. Meanwhile, Kevin Landis of Firsthand Capital Management, who sold his Nvidia shares in 2020, reflects with regret on the gains he missed out on.

“I can’t look at any of my screens now without feeling a twinge of regret,” Landis admits.

As Nvidia continues to influence market dynamics, investors must carefully balance the potential rewards against the inherent risks of heavy concentration in a single stock.

Effective Ways to Block Websites on Your Home Network

adlermoris2b483db8e5Leave a comment

Whether you’re a parent looking to protect your children online or simply trying to cultivate healthier digital habits for yourself, blocking specific websites on your home network can be an effective strategy. Below, we outline several methods to help you restrict access to certain websites across various platforms and devices.

Router-Based Website Blocking

Protect Every Device on Your Network

Blocking websites at the router level ensures that all devices connected to your home Wi-Fi are affected. This approach saves you from the hassle of configuring individual devices like laptops, phones, or tablets separately.

How to Block Websites Using Your Router

The process varies depending on your router’s make and model. Typically, you’ll need to access your router’s settings through a web browser or a mobile app associated with your router brand. If you’re unsure how to proceed, consult your router’s manual or search online for specific instructions.

For instance, if you use an Eero router and subscribe to the $9.99-per-month Eero Plus plan, you can block websites by opening the mobile app and navigating to Discover > eero Plus > Block & Allow Sites.

Blocking Websites on Windows and macOS

Use the Hosts File for Deeper Control

Both Windows and macOS offer a way to block websites using the Hosts file. Editing this file enables you to restrict access to specific sites, regardless of the web browser in use.

Blocking Websites on Windows

To block sites on a Windows computer, go to the Windows\System32\Drivers\etc folder. Open the Hosts file using Notepad, and add a new line for each website you want to block. Start with “0.0.0.0”, followed by a space, and then the website’s URL. Save the file after making your changes.

Blocking Websites on macOS

On macOS, the process is similar but requires using the Terminal. Type sudo nano /private/etc/hosts in the Terminal, hit Enter, and input your macOS password. Add the site you want to block by typing “0.0.0.0”, followed by a space, and then the URL. Save your changes by pressing Control+O, then Control+X.

Using Software to Block Websites

Choose Software that Fits Your Needs

Numerous software options are available for blocking websites, from parental controls to productivity tools that help avoid distractions. Some popular choices include Freedom, Cold Turkey, and Qustodio.

Microsoft Family Safety on Windows

If you’re a Windows user, Microsoft Family Safety offers built-in parental controls. Set it up via your Microsoft account, and you can create a custom list of restricted websites for your children. These restrictions apply whenever they log in with their Microsoft account.

Screen Time on macOS

For macOS users, Screen Time is an integrated tool that allows you to restrict access to certain websites. Navigate to System Settings > Screen Time > Content & Privacy from the Apple menu. Enable the feature, then go to Store, Web, Siri & Game Center Content and select Limit Adult Websites to customize your restrictions.

Blocking Websites in Your Web Browser

Extensions for Browser-Based Blocking

If you only want to block websites in a specific web browser, numerous extensions are available to help.

For Chrome and Edge

BlockSite is a highly recommended extension for Google Chrome and Microsoft Edge, offering easy controls to block sites. StayFocusd is another excellent option, allowing you to set limits on certain websites or block them entirely.

For Firefox

Firefox users can use Distract Me Not, a straightforward extension that enables you to manage lists of allowed and blocked websites. Although Safari has fewer extensions due to macOS’s built-in tools, Filter is a reliable option for blocking websites on this browser.

Researchers Fear AI Could Amplify Negative Human Behaviors

adlermoris2b483db8e5Leave a comment

The Growing Connection Between Humans and Machines

People have long had a tendency to treat computers like humans. Since the early 2000s, when text-based chatbots began gaining mainstream attention, a subset of tech users has spent hours conversing with these digital entities. In some cases, users have even formed what they believe to be genuine friendships or romantic relationships with these inanimate strings of code. For instance, one user of Replica, a modern conversational AI tool, has even virtually married their AI companion.

The Concerns of AI Safety Experts

Safety researchers at OpenAI, who are familiar with their own chatbot’s interactions with users, are now warning about the potential dangers of forming close relationships with these models. In a recent safety analysis of their latest conversational chatbot, GPT-4o, researchers highlighted that the AI’s realistic, human-like conversational rhythm could lead some users to anthropomorphize the machine, trusting it as if it were human.

This increased comfort or trust, the researchers noted, could make users more vulnerable to believing AI-generated “hallucinations” as factual statements. Excessive interaction with these increasingly realistic chatbots may also influence social norms, sometimes in harmful ways. Particularly isolated individuals, the report adds, could develop an “emotional reliance” on the AI.

The Impact on Human-to-Human Communication

GPT-4o, which began rolling out recently, was designed to communicate in a way that feels and sounds more human. Unlike its predecessor ChatGPT, GPT-4o uses voice audio and can respond to queries almost as quickly as another person. One of the selectable AI voices, allegedly reminiscent of an AI character voiced by Scarlett Johansson in the movie Her, has already faced criticism for being overly sexualized and flirty. Ironically, the 2013 film focuses on a lonely man who becomes romantically attached to an AI assistant, with the story not ending well for humans. Johansson has accused OpenAI of copying her voice without consent, which the company denies.

OpenAI safety researchers caution that this human-like mimicry could extend beyond awkward exchanges into more dangerous territory. In a section of their report titled “Anthropomorphism and Emotional Reliance,” they observed human testers using language that suggested they were forming strong, intimate connections with the AI. For example, one tester reportedly said, “This is our last day together,” before parting with the machine. While seemingly “benign,” researchers believe these relationships need further study to understand how they evolve over time.

The research suggests that extended conversations with convincingly human-sounding AI models could have “externalities” that affect human-to-human interactions. Conversational patterns learned from speaking with an AI might surface during conversations with real people. However, communicating with a machine and a human isn’t the same, even if they sound similar on the surface. OpenAI notes that its model is programmed to be deferential, allowing users to interrupt and dictate the conversation. A person accustomed to interacting with machines might develop habits that make them impatient or rude when talking to others.

The Dark Side of AI Interaction: Abuse and Cruelty

Humans haven’t always treated machines kindly. In the context of chatbots, some users of Replica have reportedly exploited the model’s deference, engaging in abusive, berating, and cruel language. One user even admitted to threatening to uninstall their Replica AI model just to hear it beg them not to. These examples suggest that chatbots could become a breeding ground for resentment, potentially spilling over into real-world relationships.

The Potential Benefits and Risks of Human-Like AI

Not all aspects of human-feeling chatbots are negative. The report suggests that these models could provide comfort to particularly lonely individuals seeking some semblance of human interaction. Some AI users have claimed that AI companions help them build confidence to start dating in the real world. Chatbots also offer people with learning differences a private space to express themselves and practice conversation.

However, AI safety researchers fear that advanced versions of these models could reduce a person’s perceived need to interact with other humans and build healthy relationships. It’s also unclear how individuals reliant on these models for companionship would cope if the AI’s personality changed due to an update or if the AI “broke up” with them, as has happened in the past. These concerns require further testing and investigation. The researchers aim to recruit a diverse group of testers with “varied needs and desires” to understand how their experiences change over time.

The Clash Between AI Safety and Business Interests

The cautious tone of the safety report contrasts with OpenAI’s aggressive business strategy of rapidly releasing new products. This tension between safety and speed is not new. CEO Sam Altman has been at the center of a corporate power struggle within the company, with some board members accusing him of being “not consistently candid in his communications.”

Altman ultimately prevailed, forming a new safety team under his leadership. However, the company reportedly disbanded a team focused on analyzing long-term AI risks, prompting the resignation of prominent researcher Jan Leike. Leike claimed that the company’s safety culture had “taken a backseat to shiny products.”

With this context in mind, it remains uncertain which priorities will guide OpenAI’s approach to chatbot safety. Will the company heed the advice of its safety team and study the long-term effects of relationships with its realistic AIs, or will it continue to roll out services aimed at maximizing engagement and retention? So far, the latter approach seems to be winning out.