AI companies are flush with cash, but can’t find the data they need, IBM claims a breakthrough in quantum computing and despite claiming victory, has Meta fallen behind on the development of AI?



 

These and more top tech news stories from Hashtag Trending and Tech News Day. I’m your host Jim Love, CIO of IT World Canada and Tech News Day in the US.

Generative AI startups are facing a significant challenge despite being flush with billions in funding. The issue at hand is the lack of access to the right data, a critical component for building powerful AI applications. As Brad Svrluga, co-founder and general partner of venture-capital firm Primary Venture Partners, puts it, many companies may have brilliant AI applications in mind, but without access to the right data, they can’t build a competitive product.

Venture funding in generative-AI startups has surged from $4.8 billion in 2022 to $12.7 billion in just the first five months of 2023, according to PitchBook. However, these startups are finding it challenging to access training data sets in niche areas like finance and healthcare.

Many AI startups are seeking partnerships with large enterprises that have vast amounts of data. However, these enterprises are often hesitant to share their data due to concerns about intellectual property rights and data security. Andy Baldwin, EY’s global managing partner of client service, raises questions about data ownership and access rights when data is used to train an outside model.

Startups are trying to navigate these challenges by training different models for each client using only that client’s data. However, this approach requires convincing clients of its efficacy and ensuring robust cybersecurity measures are in place to protect the data.

The struggle for data access is creating a competitive landscape where startups are racing to secure more data within certain niches. Adam Struck, founder and managing partner of Struck Capital, describes it as an “arms race” where startups are vying for exclusive access to proprietary data sets.

Sources include: The Wall Street Journal

IBM’s quantum computer, Eagle, has achieved a significant milestone by successfully simulating the behaviour of a magnetic material. This experiment, described in Nature on June 14, 2023, was able to work around quantum noise, a major hurdle in quantum computing that introduces errors in calculations. 

The IBM team used ‘error-mitigating’ techniques to perform quantum calculations “at a scale where classical computers will struggle,” according to Katie Pizzolato, who heads IBM’s quantum theory group. Although the problem they tackled was based on a simplified model of a material, the success of the experiment has sparked optimism about the potential of quantum computers to handle more complex algorithms and systems.

The experiment is seen as a benchmark in the field of quantum computing, indicating that these machines could have real-world applications within two years. IBM is also expected to unveil its most powerful processor yet, the 1,121-qubit Condor chip, later this year.

Sabrina Maniscalco, chief executive of quantum-computing start-up Algorithmiq, commented on the experiment, saying, “These machines are coming.”

Sources include: Nature 

In a recent turn of events, Reddit, the popular online platform, has found itself in the crosshairs of ransomware gang BlackCat, also known as ALPHV. The group has claimed responsibility for a phishing scheme that took place in February, which resulted in the exposure of internal documents, dashboards, code, contracts, and some information related to advertisers and employees. The data, amounting to 80GB, has not been leaked publicly yet, but BlackCat threatens to do so unless Reddit pays a ransom of $4.5 million and reverses its recent API price increases.

The phishing scheme involved “plausible sounding prompts” that led employees to a website mimicking Reddit’s intranet gateway. One employee fell for the trick, inadvertently providing the hackers with login details and second-factor tokens. The breach, however, did not compromise Reddit users’ personal information.

The ransom demand coincides with ongoing site-wide protests against Reddit’s API price hikes, which have forced popular third-party apps like Narwhal and Apollo to shut down. Apollo’s creator, Christian Selig, stated that he would need to spend $20 million per year to keep the app running under the new pricing.

Despite massive protests, with up to 8,000 subreddits going dark at one point, Reddit’s CEO, Steve Hoffman, has stood firm on the decision. He was quoted saying, “These people who are mad, they’re mad because they used to get something for free, and now it’s going to be not free.” The situation has led some advertisers to pause on the site while the blackouts continue.

BlackCat has expressed little hope that Reddit will meet its demands, stating, “We are very confident that Reddit will not pay any money for their data. We expect to leak the data.” The impact of this development on API prices remains uncertain, as Reddit has yet to comment on the situation.

Sources include: Engadget

Meta, the tech giant led by CEO Mark Zuckerberg, is facing a significant challenge in the AI landscape. Over the past year, the company has lost a third of its published AI researchers, according to the Wall Street Journal. This exodus is attributed to burnout and a lack of confidence in Meta’s direction.

In 2013, Zuckerberg brought on board Yann LeCun, an AI “godfather,” to spearhead Meta’s AI advancements. However, the company was notably absent from the White House’s recent summit of “companies at the forefront of AI innovation.” This absence, coupled with the departure of many researchers, has left Meta scrambling to catch up with industry leaders like OpenAI.

The launch of OpenAI’s ChatGPT in November has accelerated the AI innovation race, leading to even more departures from Meta. Another AI competitor, image generator Midjourney, has gained attention within a year of launching, thanks to viral fake images of Pope Francis and Donald Trump.

An internal survey conducted between April 26 and May 10 revealed that only 26 per cent of Meta employees who responded felt confident in their company’s leadership. Despite these challenges, Zuckerberg praised the company for its strides in AI during a June town hall meeting, stating, “In the last year, we’ve seen some really incredible breakthroughs—qualitative breakthroughs—on generative AI.”

Sources include: Business Insider

And that’s the top tech news stories for today. 

Links to all of the stories can be found in the text version of this podcast at itworldcanada.com/podcasts 

Hashtag Trending goes to air five days a week with a special weekend interview episode called Hashtag Trending, the Weekend Edition.  You can find us on Google, Apple, Spotify or wherever you get your podcasts. We can even be in your home on you’re your Alexa or Google smart speaker. 

We’re also on YouTube five days a week with a video newscast only there we are called Tech News Day. Check us out. Give us a like.

We love to hear your comments.  You can find me on Linked In, Twitter or on our new Mastodon site technews.social where I’m @therealjimlove.

Or if that’s too much to remember, just go to the article at itworldcanada.com/podcasts and you’ll find a text version with additional links and references.  Click on the x or the check mark and tell us what you think. 

Oh, and one more thing.  Have a terrific Tuesday!

The post Hashtag Trending Jun.20-AI startups lack data despite flush with cash; IBM claims breakthrough with quantum computing; Is Meta falling behind in AI development? first appeared on IT World Canada.

Leave a Reply