Menu
Menu

Assessing ChatGPT’s Reliability in Drug Information

Avatar photo Jerry Cooper on December 5, 2023

In the ever-evolving landscape of artificial intelligence, ChatGPT has emerged as a prominent player, capturing the attention of users worldwide. However, recent scrutiny has been cast upon the accuracy of its responses, particularly in the realm of drug-related inquiries. This examination delves into the intricacies of ChatGPT’s performance, as revealed in a study conducted by pharmacists at Long Island University.

ChatGPT Study Results:

A thorough exploration of the study’s findings illuminates a concerning pattern. Out of 39 drug-related questions posed to the free version of ChatGPT, a mere 10 responses were deemed “satisfactory” based on established criteria. The remaining 29 questions elicited responses that were either indirect, inaccurate, or incomplete.

Methodology:

The study drew from real questions addressed to Long Island University’s College of Pharmacy drug information service, spanning from January 2022 to April of the current year. Pharmacists meticulously researched and answered 45 questions, serving as the benchmark against which ChatGPT’s responses were measured.

Key Findings:

    1. Incomplete and Inaccurate Responses: ChatGPT fell short in directly addressing 11 questions, provided inaccurate responses to 10 questions, and offered incomplete or incorrect answers to another 12.
    2. Lack of References:  Despite the researchers’ explicit request for references in each response, ChatGPT provided references in only eight responses, each citing sources that do not exist.
  • Case Study Examples:
  • Drug Interaction: An alarming instance revealed that ChatGPT inaccurately indicated no reported interactions between Pfizer’s Covid antiviral pill Paxlovid and the blood-pressure-lowering medication verapamil, potentially exposing patients to preventable side effects.
  • Dose Conversion: When tasked with guiding the conversion of doses between two forms of the drug baclofen, ChatGPT’s response included an unsupported conversion method, coupled with a critical error in displaying intrathecal doses in milligrams instead of micrograms.

Recommendations:

Lead author Sara Grossman underscores the imperative for caution. Users, including healthcare professionals and patients, are advised to exercise due diligence by corroborating ChatGPT’s responses with information from trusted sources, such as medical professionals or authoritative government-based medication information websites like the National Institutes of Health’s MedlinePlus.

Concerns and Limitations:

Grossman, the study’s lead author, acknowledges that the examination focused exclusively on the free iteration of ChatGPT. While recognizing the possibility of divergent results with a paid version, the study intentionally replicated the experience of the general user population, which predominantly engages with the free version.

Takeaway:

This comprehensive analysis serves as a poignant reminder of the critical importance of prudence when relying on ChatGPT for drug-related information. It advocates for users to supplement ChatGPT’s responses with information from reliable sources, highlighting the need for continual scrutiny and enhancement in the accuracy of medical information disseminated by AI chatbots.

Resilience Prevails: ChatGPT Bounces Back from a Major Outage

Avatar photo Alice Blake on December 4, 2023

The ChatGPT Outage and its Impact

In a remarkable turn of events, OpenAI’s celebrated chatbot, ChatGPT, encountered a brief yet notable disruption in its service, highlighting both the challenges and the tenacity that underlie the world of AI-powered communication. This incident took place on a typical Wednesday morning when ChatGPT began to display unexpected errors, leading to a temporary but impactful outage. The ramifications were not confined solely to ChatGPT itself but rippled through OpenAI’s entire ecosystem, affecting more than two million developers who rely on the platform’s API services for a myriad of applications.

For ChatGPT users, the experience was marked by the disheartening message, “ChatGPT is at capacity right now.” This simple statement served as a stark reminder of the complexities and demands associated with handling an AI-driven platform of such magnitude. Consequently, OpenAI promptly addressed this situation by categorizing it as a “major outage,” recognizing its gravity.

The Swift Resolution and Recovery

In response to the unexpected outage, OpenAI’s approach was nothing short of commendable. Their dedicated team worked tirelessly to identify the root cause of the problem and promptly implemented a solution. This swift response meant that the disruption was relatively short-lived, with services returning to their usual operational state. This incident showcased the resilience of ChatGPT’s underlying infrastructure and the commitment of the team behind it to ensure uninterrupted access to the platform’s capabilities.

A Glimpse into ChatGPT’s Expanding Influence

Beyond the technical hiccup that briefly eclipsed ChatGPT’s performance, a broader narrative is emerging regarding the growing influence of this AI-driven communication tool. This incident occurred in the context of OpenAI’s recent unveiling of GPT-4 Turbo, a cutting-edge AI model that promises even more advanced capabilities and performance. The release of this enhanced AI model further underscores OpenAI’s commitment to pushing the boundaries of what AI can achieve.

Additionally, OpenAI’s decision to allow users to create customized versions of ChatGPT represents a significant step forward in tailoring AI-driven interactions to meet specific needs. The impact of this customization feature on various industries is yet to be fully realized, but it is indicative of the evolving landscape of AI applications.

Furthermore, statistics shared by OpenAI’s Chief Technology Officer, Mira Murati, during the company’s first in-person event, provide a glimpse into ChatGPT’s expanding influence. More than 92% of Fortune 500 companies now leverage the platform’s capabilities, a substantial increase from the 80% reported just a few months ago. These companies span a wide range of sectors, from financial services to legal applications and education, highlighting the versatility and relevance of ChatGPT in today’s business landscape.

Morgan Stanley Embarks on the Generative AI Revolution in Finance

Avatar photo Jerry Cooper on December 4, 2023

I. Morgan Stanley and AI

Morgan Stanley, a prominent player on Wall Street, has ushered in the era of generative AI, marking a significant milestone in the financial industry’s evolution. In a groundbreaking move, the bank has introduced an AI-powered assistant, created in collaboration with OpenAI’s cutting-edge generative AI software, which is now fully operational and available to all financial advisors and their support teams.

II. A Transformational Leap for Financial Advisors

Morgan Stanley’s co-President, Andy Saperstein, underscored the pivotal role of financial advisors, emphasizing that they remain at the heart of Morgan Stanley’s wealth management efforts. Moreover, he articulated the bank’s belief in the transformative potential of generative AI, foreseeing a revolution in client interactions, increased efficiency in advisor practices, and the liberation of valuable time for advisors to focus on their core mission: serving clients effectively.

III. Pioneering the Use of GPT-4 on Wall Street

Morgan Stanley, renowned as a powerhouse in investment banking and wealth management, made waves when it revealed its collaboration with OpenAI’s GPT-4. While competitors like Goldman Sachs and JPMorgan Chase have also ventured into projects involving generative AI technology, Morgan Stanley stands out as the first major Wall Street firm to deploy a bespoke solution based on GPT-4 for its employees. According to Jeff McMillan, Head of Analytics, Data, and Innovation at Morgan Stanley wealth management, this innovative tool bears the name “AI @ Morgan Stanley Assistant.”

IV. Access to Intellectual Capital

The AI @ Morgan Stanley Assistant is a game-changer, offering financial advisors swift access to the bank’s extensive “intellectual capital” repository, containing approximately 100,000 research reports and documents. This strategic advantage streamlines responses to queries about markets, recommendations, and internal procedures, allowing advisors to dedicate more time and attention to client engagement.

V. Humanizing Interactions with AI

Despite its seemingly simple text-based interface, the AI @ Morgan Stanley Assistant conceals the complexity of ensuring high-quality responses. Months of meticulous curation of documents and rigorous testing by human experts were essential to guarantee the program’s efficacy. Notably, advisors will need to frame their questions in complete sentences, mirroring human conversation, rather than relying on keywords, as is common with search engine queries—a shift that may require some adjustment.

VI. A Glimpse into the Future

The introduction of the AI @ Morgan Stanley Assistant represents just the initial step in a series of generative AI solutions planned by the bank. Morgan Stanley is currently piloting a tool known as “Debrief,” which automates the summarization of client meeting content and generates follow-up emails—a potential game-changer in enhancing client relationships and workflow efficiency.

VII. Embracing a New Paradigm in AI

Adopting OpenAI’s software necessitated a fundamental departure from traditional coding approaches. OpenAI’s ChatGPT leverages large language models (LLMs) to produce human-like responses to inquiries. Rather than crafting code in the conventional sense, this new approach involves presenting examples of “good” responses, allowing the system to learn and apply logic akin to human reasoning—a shift that underscores the transformative nature of AI in financial services.

VIII. A Disruptive Force with Unprecedented Opportunities

The excitement surrounding AI has not only impacted the financial sector but has also driven the broader stock market and compelled entire industries to grapple with its far-reaching implications. Some experts have even posited AI as the next foundational technology. Jeff McMillan, with two decades of experience in artificial intelligence, expressed his awe at the current landscape, describing it as “completely disruptive.” Morgan Stanley recognized this window of opportunity and took proactive steps to ensure it remains at the forefront of this groundbreaking transformation.

IX. Takeaway on Morgan Stanley and AI

Morgan Stanley’s foray into the generative AI era marks a pivotal moment in the financial industry’s evolution. With the AI @ Morgan Stanley Assistant, the bank is not merely embracing innovation but also reshaping the way financial advisors interact with data, clients, and the future of wealth management.

AI Milestones: A Week of Surprises and Innovations

Avatar photo Jerry Cooper on December 4, 2023

Last week witnessed a series of groundbreaking events in the realm of artificial intelligence, setting the stage for significant advancements and sparking discussions across the industry. From OpenAI’s major announcements to Elon Musk’s unveiling of Grok, and the debut of Humane’s Ai Pin, the week was nothing short of monumental.

OpenAI’s Strides in AI Development:

In a move reminiscent of an arms race, OpenAI hosted its inaugural developer conference, marking a pivotal moment since the launch of ChatGPT. The conference showcased a range of updates, including the introduction of GPTs, customizable versions of ChatGPT. These GPTs, akin to plugins, promise versatile applications such as database connectivity, email integration, and e-commerce facilitation. With a strategic focus on developers, OpenAI aims to address commercial challenges by fostering a robust developer ecosystem.

GPT Store Launch:

To further enhance accessibility, OpenAI announced the upcoming launch of the GPT Store. This platform will allow users to search and discover GPTs, categorizing them based on utility across productivity, education, and entertainment. The move reflects OpenAI’s commitment to providing a diverse array of tools to meet varied user needs.

GPT-4 Turbo Unveiled:

CEO Sam Altman introduced GPT-4 Turbo, the latest iteration powering ChatGPT. Boasting enhanced capabilities, GPT-4 Turbo can process input equivalent to about 300 pages of a standard book—significantly surpassing its predecessor. Altman’s revelation of the platform’s growth, with 2 million developers and 100 million active users, underscores the widespread adoption of OpenAI’s technology.

Humane’s Revolutionary Ai Pin:

Humane, a startup founded by former Apple employees, presented a glimpse into the future with the introduction of Ai Pin. This AI wearable, equipped with a Snapdragon processor and Qualcomm AI engine, aims to redefine human-machine interaction. Packed with depth and motion sensors, an ultra-wide camera, and a laser ink display, Ai Pin envisions replacing smartphones by projecting information onto users’ hands. Despite the promising features, industry analysts raise concerns about potential challenges, including battery life and societal acceptance of always-on AI.

Grok’s Arrival on X:

Elon Musk’s xAI took inspiration from literary humor, introducing Grok—a chatbot with a sarcastic sense of humor, reminiscent of Musk himself. While still in early testing, Grok is set to join X’s Premium+ service, offering a unique blend of wit and real-time access to information. Musk’s move into the chatbot space adds a new dimension to AI applications on social platforms.

OpenAI’s Resilience Amidst Challenges:

Two days post-conference, OpenAI faced disruptions attributed to potential targeted attacks on its servers. The distributed denial of service (DDoS) attack led to periodic outages, showcasing the vulnerability of AI infrastructure. OpenAI, however, assured users that no sensitive information was compromised, highlighting the need for robust security measures in the evolving landscape of AI.

AI Landscape Beyond OpenAI:

Major players in the tech industry continue to invest heavily in AI. Amazon’s commitment to training “Olympus,” an AI codenamed project with double the parameters of OpenAI’s GPT-4, underscores the competitive race in AI development. Additionally, YouTube is exploring AI tools to enhance user interactions, from answering questions to providing content recommendations.

Takeaway:

The events of the past week underscore the dynamic nature of the AI landscape. As OpenAI pushes the boundaries of AI capabilities, other tech giants follow suit, contributing to an era of rapid innovation. While challenges such as security and societal acceptance persist, the relentless pursuit of advancements ensures that the AI journey is one of constant evolution.

The Transformational Impact of GenAI in Economics

Avatar photo Jeah Lorraine Adams on December 4, 2023

GenAI, or generative artificial intelligence, is poised to revolutionize the field of economics, as highlighted by Anton Korinek, a prominent economics professor at the University of Virginia. This groundbreaking technology has the potential to transform economic research, teaching, and forecasting. This article explores the various dimensions of GenAI’s influence on the field of economics and its implications for economists.

GenAI: A Game-Changer in Economics

Economists have long been utilizing machine learning for data analysis and economic projections. However, GenAI represents a distinct technology, underpinning tools like ChatGPT and advancing at an astonishing pace. According to Korinek, it’s set to “revolutionize research.”

Elevating Economic Research

Large language models, a specific type of GenAI, can aid economists in multiple ways, from generating research ideas to evaluating research plans, as well as assisting with writing, copy editing, and text summarization. They can enhance the clarity and specificity of economic research.

The Power of AI Chatbots

Prominent GenAI tools, such as ChatGPT, New Bing, Bard, Claude 2, and LlaMA 2, are instrumental in brainstorming, copy editing, and summarizing text. They offer unprecedented capabilities for improving writing, suggesting titles, and simplifying social media promotion. Moreover, AI chatbots are adept at summarizing lengthy academic research papers, making them invaluable to economists.

Technical Tasks Made Easier

Economic research often involves technical tasks like coding and mathematical proofs. GenAI tools can assist in writing, explaining, translating, and debugging code, especially in languages like Python and R. They can set up economic models, derive equations, and provide explanations, although their mathematical capabilities are currently limited.

The Evolution of GenAI

It’s crucial to note that the latest versions of GenAI chatbots, like ChatGPT-4, offer enhanced capabilities compared to their predecessors.

GenAI’s Limitations

While GenAI holds tremendous promise, it’s not infallible. It can occasionally produce inaccurate information, referred to as “hallucinations.”

A Valuable Teaching Assistant

Economics professors at George Mason University have shown that GenAI can be highly beneficial in the classroom. It can summarize text, improve writing, suggest ideas, and solve simple economic models with explanations. This makes it a powerful teaching assistant for economics courses.

Forecasting with Precision

Recent studies indicate that GenAI is exceptionally skilled at forecasting inflation, outperforming human economists in some cases. This has significant implications for macroeconomic predictions and economic policy.

The Impact on Employment

The advancement of GenAI is expected to have a limited impact on employment initially. It will primarily enhance economists’ productivity and efficiency. However, it may eventually lead to job losses, particularly in fields that involve repetitive tasks. The risk to employment is a subject of concern and requires careful consideration.

Takeaway:

GenAI’s influence on the field of economics is undeniable. It promises to elevate research, teaching, and forecasting, providing economists with powerful tools to enhance their work. However, it’s essential to recognize the technology’s limitations and its potential impact on employment.

ChatGPT dipped in quality according to a new study

Avatar photo Scarlet Logan on September 5, 2023

ChatGPT — With the debut of ChatGPT, a novel AI language model, in late 2022, OpenAI grabbed the globe by storm. The success of the AI service cleared the door for a one-of-a-kind AI race, with hundreds of tech companies vying to imitate it.

While the service has received some criticism, OpenAI has taken the liberty of upgrading it, honing the language model to be as faultless as possible. ChatGPT seems to have found its stride after a couple of revisions.

The most current version of the AI language model pioneer sparked a coin rise, sparking demands to halt development. A new research, however, seems to show that the AI bots may have suffered a setback, leading to a decrease.

Read also: Bitcoin mining claims sparks response from CH4 Capital co-founder

The ChatGPT study 

Between March and June 2022, researchers from Stanford and UC Berkeley conducted a study in which they rigorously analyzed multiple versions of ChatGPT. They devised stringent criteria to assess the chatbot’s proficiency in coding, arithmetic, and visual thinking tasks. The outcome of ChatGPT’s performance was not favorable.

According to the results of the testing, there was a worrying reduction in performance between the versions tested. ChatGPT answered 488 of 500 questions correctly during a math challenge regarding prime numbers in March, yielding a 97.6% accuracy rate. By June, the percentage had decreased to 2.4%, with only 12 questions correctly answered.

The deterioration did not become obvious until the chatbot’s software development talents were examined.

“For GPT-4, the percentage of generations that are directly executable dropped from 52.0% in March to 10.0% in June,” the study said.

The findings were discovered using the models’ pure versions, which did not include any code interpreter plugins.

When it came to reasoning, the researchers used visual prompts and a dataset from the Abstract Reasoning Corpus. There was a clear drop, although it wasn’t as high as in math and coding.

“GPT-4 in June made mistakes on queries on which it was correct for in March,” the study said.

Possible reasons for the decline

The drop was unexpected, prompting the question, “What could explain ChatGPT’s painfully obvious downgrades in recent months?” According to a proposed hypothesis, researchers speculate that it could be a side effect of OpenAI’s optimizations.

Another plausible reason is that the adjustments were implemented as a precaution to prevent ChatGPT from responding to harmful inquiries. However, the safety alignment may limit ChatGPT’s use for other activities.

The model, according to the researchers, has a propensity to offer wordy, indirect solutions rather than unambiguous ones.

On Twitter, AI researcher Santiago Valderrama commented, “GPT-4 is getting worse over time, not better.” He also hinted that a cheaper, quicker mix of models may have replaced the original ChatGPT framework.

“Rumors suggest they are using several smaller and specialized GPT-4 models that act similarly to a large model but are less expensive to run,” he noted. 

Valderrama also stated that using smaller models might result in speedier answers, but at the expense of less expertise.

“There are hundreds (maybe thousands already?) of replies from people saying they have noticed the degradation in quality,” Valderrama continued. “Browse the comments, and you’ll read about many situations where GPT-4 is not working as before.”

Other insights

After trying to make sense of the data, Dr. Jim Fan, another AI researcher, commented on some of his observations on Twitter. Fan related them to how OpenAI refined its models.

“Unfortunately, more safety typically comes at the cost of less usefulness, leading to a possible degrade in cognitive skills,” he wrote.

“My guess (no evidence, just speculation) is that OpenAI spent the majority of efforts doing lobotomy from March to June, and didn’t have time to fully recover the other capabilities that matter.”

Fan also pointed out that the safety alignment made code needlessly lengthy, mixing in irrelevant material regardless of the prompts.

“I believe this is a side effect of safety alignment,” he offered. “We’ve all seen GPTs add warnings, disclaimers, and back-pedaling.”

Fans said that cost-cutting measures, as well as the introduction of warnings and disclaimers, might have contributed to ChatGPT’s collapse. Furthermore, the absence of widespread community feedback might have been a role. Although more testing is required, the end results corroborated users’ concerns about the diminishing coherence of ChatGPT’s once-highly lauded outputs.

To avoid future deterioration, enthusiasts have advocated for open-source models such as Meta’s LLaMA, which allows for community debugging. They also stressed the need of constant benchmarking in detecting regressions.

Meanwhile, ChatGPT fans should moderate their expectations because the unique and groundbreaking language model AI chatbot appears to have degraded in quality.

Bard is unveiled, Alphabet shares receive a boost

Avatar photo Alice Blake on March 23, 2023

…

Bard – The race for AI has taken another turn as Google’s efforts have finally paid off, and now the company is eager to launch its chatbot service.

The initial backlash the company faced a couple of weeks earlier seems to have been drowned out with the company’s latest foray into the AI space.

Soon, chatbot enthusiasts can utilize Google’s latest innovation: Bard.

The news

Google recently announced opening access to its AI chatbot tool Bard, the company’s answer to rival ChatGPT.

Since Tuesday, users have been given access to a waitlist, allowing them to experiment with Bard.

Bard has similar tools that can allow users to try the following and more:

  • Outline and create essay drafts
  • Write out a plan for a baby shower
  • Create lunch ideas with the content of a refrigerator

According to Google, it will start rolling out Bard to the United States and the United Kingdom, with plans to expand the tool to more countries and languages in the future.

AI extravaganza

The news comes after several major tech companies, like Google, Microsoft, and Facebook, among countless others, stepped up their efforts in the race to develop and deploy AI-powered tools.

The increased focus on AI sprang up after the viral success of OpenAI’s ChatGPT.

Last week, Google announced it would integrate AI into its productivity tools, such as Gmail, Google Docs, and Google Sheets.

After their news, Microsoft announced a similar upgrade to its productivity tools.

In February, Google unveiled Bard through a demo that sparked criticism after it provided an inaccurate response to some questions.

As a result, Alphabet, Google’s parent company, experienced a 7.7% share drop, wiping out $100 billion from its market value.

Read also: Baidu stocks improve after ERNIE Bot demo

ChatGPT impact

Bard follows ChatGPT’s model, wherein it is built on a massive language model.

ChatGPT, a product developed by AI research firm OpenAI, was released to the public in late November.

The models are trained on massive amounts of data online to help the AI create unique responses to the user’s creative prompts.

It was ChatGPT’s success and the spotlight given to it that led to Google management calling a “code red” situation for its search business.

However, Bard’s mistakes during the demo emphasized the challenge Google and other companies had with integrating AI technology into their core products.

Large language models are helpful, but they can also present several issues, including:

  • Biases
  • Incorrect facts
  • Responding aggressively

On Tuesday, Google released a blog post saying its AI tools are stall prone to mistakes.

The company reassured people that it still uses human feedback to improve its system and add guardrails, such as capping the number of exchanges in a dialogue to keep the interactions helpful and stay on topic.

Last week, GPT-4 was launched as the next-generation version of the tech and to power Microsoft’s Bing browser with similar safeguards.

After its first day, GPT-4 surprised users with its early test and company demo.

Public sway

Sundar Pichai, the CEO of Google and Alphabet, told employees that Bard’s success would largely rely on public testing in an email.

“As more people start to use Bard and test its capabilities, they’ll surprise us. Things will go wrong,” Pichai wrote.

“But the user feedback is critical to improving the product and the underlying technology.”

The message came as Google launched Bard.

Following the announcement, Alphabet shares were up nearly 4% in mid-day trading.

Pichai’s email also revealed that over 80,000 Google employees helped update Bard’s development following his all-hands-on-deck call to action in February.

The Tuesday note also said the company is working on testing responsibly, having invited 10,000 testers from different backgrounds and perspectives.

Pichai also told employees they should be proud of their work and the years of tech breakthroughs that led them to where they are.

“Even after all this progress, we’re still in the early stages of a long AI journey,” he said.

“For now, I’m excited to see how Bard sparks more creativity and curiosity in the people who use it.”

Baidu stocks improve after ERNIE Bot demo

Avatar photo Scarlet Logan on March 20, 2023

…

Baidu – Chinese search giant Baidu received a massive share bump after it revealed its answer to the ChatGPT trend: ERNIE Bot.

On Friday in Hong Kong, the stock surged 14.3%, making it the top company in the Hang Seng Index.

In addition, the company gained 3.8% in New York during US trading hours on Thursday.

What happened?

A day earlier, Baidu was the most prominent loser in the Hang Seng Index.

A public demonstration of the company’s bot left investors unimpressed, leading Hong Kong shares to drop 6.4%.

However, CEO Robin Li revealed during the presentation that more than 650 companies had joined the ERNIE ecosystem since February.

The reversal follows the company’s announcement of more than 30,000 businesses signing up to try the chatbot service two hours after the demonstration.

Esme Pau, the head of China and Hong Kong internet and digital assets for Macquarie, said:

“The high degree of enterprise interest is positive, and we expect Baidu to continue to capture China’s enterprise demand for generative AI.”

According to Pau, the company’s shares bounced back on Friday following positive feedback from users and analysts.

The reviews suggested the bot was highly advanced.

Stock drop

During the company’s presentation, Baidu showed its chatbot’s capabilities, like generating a company newsletter, creating a corporate slogan, and solving a math riddle.

According to Esme Pau, the company’s stocks dropped on Thursday due to the demo being pre-recorded.

Because it wasn’t live, investors were skeptical of the authenticity of the ERNIE Bot.

Furthermore, Pai pointed out Baidu’s demo came days after GPT-4 was launched, raising the bar for ERNIE.

Read also: Apple thieves target passcodes before snatching iPhones

GPT-4

OpenAI launched GPT-4 as the latest update to its artificial intelligence technology, receiving wide acclaim from users.

Many were stunned by early tests and a company demo showcasing its ability to do the following:

  • Draft lawsuits
  • Pass standardized exams
  • Build a working website from hand-drawn sketches

On Tuesday, OpenAI introduced GPT-4 to change how the internet is used for work, play, and creating content.

The latest update is trained on online data to create unique responses to user prompts.

The update allows users to perform the following with ease:

  • Analyze beyond texts
  • Made coding easier
  • Pass tests
  • Provide more accurate responses
  • Streamline work across various industries

Despite its advancement, OpenAI said GPT-4 still has similar limitations to previous versions.

For example, the technology is limited to its data set, which cuts off in September 2021.

ERNIE

Esme Pau said Baidu’s shares were modestly down before its showcase on Thursday, indicating pressure from investors with high expectations after the GPT-4 launch.

“ERNIE also does not have the [same] multilingual capability as GPT-4, and has yet to improve for English queries,” said Pau.

“Also, the ERNIE launch did not provide sufficient quantifiable metrics compared to the GPT-4 launch earlier this week.”

ERNIE is similar to ChatGPT in that the technology is based on a language model trained on a massive amount of online data to create unique responses to user prompts.

Robin Li said Baidu expects ERNIE to be close to ChatGPT or GPT-4.

However, he acknowledged the software has yet to be perfected, noting that it was launched to enterprise users first.

Baidu’s latest service isn’t available to the public yet.

Competition

Baidu announced its chatbot development in February.

At the time, critics said the service would only add to the current US-China rivalry regarding technology and innovation.

However, Li shook off the comparison over the launch, saying:

“The bot is not a tool for the confrontation between China and the United States in science and technology, but a product of generations of Baidu technicians chasing the dream of changing the world with technology.”

“It is a brand new platform for us to serve hundreds of millions of users and empower thousands of industries.”

According to Baidu, the company’s service stands out due to its advanced grasp of Chinese queries and ability to create different kinds of responses.

“ERNIE Bot can produce text, images, audio, and even video given a text prompt,” the company said in its statement.

“And [it] is even capable of delivering voice in several dialects, such as the Sichuan dialect.”

However, Baidu is not alone in the development of similar technology in China.

In February, Alibaba announced plans to launch a ChatGPT-style tool.

According to analysts, Baidu has the best advantage in the space in China so far.

“Our view is ERNIE is six months ahead of its potential contenders,” said Esme Pau.

On Friday, Baidu announced a milestone in its transportation business, becoming the first Beijing operator to provide fully driverless ride-hailing services.

Despite its innovation, the company isn’t allowed to charge passengers in the capital yet.

It was previously required to have a driver in the front passenger seat to assume control in emergencies.

 

Publishers wary of chatbots & ChatGPT, set for new direction

Avatar photo Jeah Lorraine Adams on February 27, 2023

…

Publishers – AI chatbots have gained popularity in recent months due to their innovative nature and ability to simplify work and learning.

Although they are helpful to the majority of people, they might represent a risk to publishers.

A new rival

Generative AI chatbots may not only produce text but also answer inquiries.

OpenAI’s ChatGPT and Microsoft’s upcoming ChatGPT-powered Bing may steal the audience that generates search-driven traffic for publishers.

Several lifestyle publications have abandoned SEO-driven content due to the competition caused by technology.

In contrast, Bustle Digital Group and Leaf Group develop original material based on personal experiences and viewpoints.

Wesley Bonner, BDG’s head of social and audience growth, claimed that the company’s editorial shift will focus on original visual material.

According to Bonner, they would prioritize the creation of amusing stories that are related to common situations and offer some advice.

Likewise, Hunker, Leaf Group’s home design website, has said that its content would focus on the writers’ viewpoints, knowledge, and ideas.

Traffic

The decision by the publishers to shift course marks a considerable investment.

Other lifestyle publishers, such as BDG, Leaf Group, and Trusted Media Brands, have not seen a notable increase in search traffic since the launch of ChatGPT in November.

Trusted Media Brands earn 80% of their referral traffic from search, whereas BDG receives 25% to 30%.

According to Beth Tomkiw, chief content officer of Trusted Media Brands, AI chatbots taking over Google search will be a bigger problem.

“My hope is that there will still be a place – even if it’s a smaller place – for the quality of work that comes from a real human,” said Tomkiw.

She is discussing what may have happened to TMB’s editorial approach if things had changed.

While reducing search-driven content is not a new issue for publishers, it is gaining popularity.

Historically, publishers use a scale approach to attract an audience, but it seldom works for firms that have already demonstrated their success.

Instead, publishers have worked to create connections with their customers during the previous decade, relying less on referral traffic via subscriptions and newsletters.

“For publishers who are still very focused on the page view as a primary metric, that’s going to be a bit of a problem,” said Jim Robinson, Clickseed’s founder.

Clickseed is an SEO and audience growth service that focuses on publishers.

“If that’s been your strategy, you might be a little behind the curve anyway.”

Read also: Google’s new focus is AI after ChatGPT pressure

Shifting from SEO

People’s search habits have already been changed by ChatGPT and AI chatbots.

According to Emma Rosenblum, BDG’s chief content officer, the firm is committed to shifting its traffic strategy away from SEO-based pieces and quick news bursts.

Rosenblum noted that online media companies were created on the low-hanging fruit of service tales that would most likely be outmoded in the next five years.

She stated that technology advancement will make it easier, quicker, and less expensive than hiring experienced writers.

“We don’t want to be doing those stories,” said Rosenblum. “That utility that we provide is going to disappear so quickly.”

“[And] I’m glad because we hate doing stuff like that. All the things that a computer could not replicate is where we’re going to put our money.”

Rosenblum claims that the business is investing in original visual content, interviews, profiles, and feature articles.

As a consequence, they will produce fewer social media posts and more short-form films.

Melissa Chowning, the founder and CEO of Twenty-First Digital, noted that now that ChatGPT has established itself, photography and visuals are the most significant assets for lifestyle magazines.

Rosenblum sent out an email outlining BDG’s strategy for compensating for the anticipated loss of traffic.

“If traffic dips a bit, it dips,” the email said.

“Chasing Google is a losing war for digital media companies, which is why we’re building up areas of our business like events and newsletters, neither of which are dependent on outside platforms.”

Furthermore, BDG’s newsletter business grew by 32% year on year, reaching 5 million subscribers.

With less traffic, the advertisements will almost surely have an effect on BDG.

Yet, Rosenblum noted that programmatic revenue would continue to be a part of their business in the future, with funds coming through direct advertising.

“In this new world, we’re expecting our revenue from events and newsletters to grow enormously, offsetting any potential programmatic loss,” she added.

Chatbot adoption

The ChatGPT launch, according to Eve Epstein of Leaf Group, is a continuation of Google search’s evolution, which is nothing new.

Publishers had to deal with a “featured snippet” in 2014, which took a section of a publisher’s website and used it to answer a user’s inquiry on Google search.

Nonetheless, it remains to be seen how things will turn out with the usage of chatbots.

According to Robinson, it is too early for publishers to implement significant strategic adjustments.

He recommended that publishers monitor their referral traffic statistics for any changes in user behavior.

“I think there is an immediate need to be having these discussions,” said Robinson.

“That plan is a good one anyway, even if you take ChatGPT out of the picture. Who wants to give all that power to Google?”

Google shares dropped from presentation

Avatar photo Jeah Lorraine Adams on February 16, 2023

…

Google – The race for AI technology has intensified since ChatGPT unveiled OpenAI in late 2022, leaving other tech firms in the dust.

Google in particular is lagging and has been working to catch up.

The company held an event on Wednesday to display Bard, an AI chatbot, to terrible consequences.

As a result, Alphabet, the parent company of Google, saw a decline of more than 7% in share price at the close of trade.

The news

On Tuesday, Microsoft showcased brand-new AI technologies on its Bing search engine.

Due to the event’s success, Google decided to emulate it.

Earlier that day, Google had confirmed the news of its Bard announcement and said that the AI technology will be made available over the coming weeks.

The presentation

Google executives spoke about Bard’s potential on Wednesday at the event.

In a presentation, the pros and cons of AI were discussed.

The company’s well-known language model, LaMDA (Language Model for Dialogue Applications), drives Bard.

Google said in a blog post on Monday that “trusted testers” will have access to chat technology before it is made more widely available.

Throughout the event, the company demonstrated upgrades to other products, such as Maps and Google Lens.

Despite its demonstration, Alphabet shares fell because investors had high expectations given Microsoft’s growing competitiveness.

AI update

On Tuesday, Microsoft’s Redmond, Washington, headquarters hosted an AI conference.

The event’s objective was to showcase AI-powered upgrades to Microsoft’s Bing and Edge browsers.

Bing has always trailed behind Google in terms of search engine usage, but advancements in AI may close the gap with conversational replies to inquiries.

Microsoft invested enormous sums in ChatGPT’s OpenAI technology, which served as the foundation for the advancements made to its products.

Read also: Google’s new focus is AI after ChatGPT pressure

ChatGPT

ChatGPT is the name of the artificial intelligence software that is making waves online.

After its November release, it generated viral content in accordance to the given instructions.

But some analysts and Google employees are starting to question if the top search engine is falling behind in AI.

The company has also been focusing on AI for a long time.

After ChatGPT’s meteoric ascent to stardom, Google instituted an internal “code red” in an effort to hasten the creation of Bard and other products.

Additionally, after years away from the day-to-day management of the company, Google co-founders Larry Page and Sergey Brin decided to take control.

Microsoft gains

Although Google has been under more pressure as a result of Microsoft’s recent AI developments, many believe it will still be some time before Microsoft sees tangible advantages.

Brent Thill, a Jeffries analyst, said the following in a note on Tuesday:

“Search improvements will act as a tailwind to [advertising revenue long term], but it will take time to bring users back to Bing, and they will need a crowbar to pry away advertisers from Google.”

“We view these updates as the tip of the iceberg for MSFT’s AI capabilities, with the largest opportunities in enterprise use cases.”

The news that was presented during the Google event, according to Evercore analysts, would have progressed the company.

Stock drop

The lack of an increase may have caused the company’s stock price to decline.

Analysts assert that they believe the incident was a hurried and likely early demonstration of the artificial intelligence that Google has been developing for years.

Many believe Google’s AI technology is strong enough to compete on its own despite these limitations.

On Wednesday, analysts released a report that said:

“Leveraging its years of AI investment (which drove a near doubling of CapEx in 2018) and unparalleled scale, this should help the company defend its market position in the long run.”

 

Menu
Menu
Menu
Menu
Menu
Menu

© 2023 Kivo Daily. All Rights Reserved