OpenAI Slashes Prices On 2 AI Models As Businesses Fight High Costs

Message received. OpenAI's announcement of dramatic price cuts for some of its AI models suggests that business resistance to AI pricing now threatens its growth so severely that OpenAI recognizes it has to take drastic measures. 

This is one of a number of signals that Silicon Valley, in its haste, greed and over-hyping has wounded, if not yet killed, the AI golden goose. From widespread public rejection of data centers (support for the centers has now become a serious political liability) to investors' skepticism about tech financial projections to corporate refusal to pay unsupportable prices, the AI FOMO has largely evaporated. The larger questions are what this means for its future growth trajectory, ability to provide returns to investors and its impact on the broader economy. JL

Cris Tolomia reports in Quartz:

OpenAI announced it is reducing the price of two models in its GPT-5.6 family - cutting the cost of GPT-5.6 Luna by 80% and GPT-5.6 Terra by 20% - three weeks after the models launched. The cuts arrive as enterprise customers have grown reluctant to authorize large AI expenditures without a clear picture of returns. Cheaper Chinese open-source models have sharpened the competitive environment, pushing OpenAI and Anthropic to make the case that their products are worth the premium. The broader backdrop has been building for some time. Enterprise customers have been cutting AI spending and demanding clearer returns on investment. OpenAI attributed the reductions to efficiency gains made during internal development of GPT-5.6.

Russians Try Armored Assaults Again - And Reap Same Failure

If repeating the same failed action and expecting a different outcome is the definition of insanity, the Russian military needs to be committed. 

Under constant pressure from the Kremlin to continue offensive operations, Russian commanders find themselves recycling old tactics in place of new tactics in hopes that something might eventually work. At Dobropillia, the Russians launched an armored assault on Ukrainian positions, evidently in hopes that the Ukrainians would be expecting another infiltration attempt by small numbers of Russian infantry (which was the tactic that replace armored assaults...) and so would miss the approach of several dozen tanks, infantry fighting vehicles and other weaponry. As anyone rational can imagine, the Ukrainians did not miss the incoming armor. The attack was annihilated. JL

Vikran Mittal reports in Forbes:

Russian forces launched their largest mechanized assault in many months against Ukrainian positions in the Donetsk Oblast. The attack was part of a larger, multi-axis offensive around Shakhove. The attack failed. Ukraine employed FPV drones, heavy bomber drones, and artillery. The Ukrainians had laid minefields along the likely avenues of approach and damaged bridges leading toward their positions. Russian attempts to use pontoon bridges, were destroyed by Ukrainian aviation. Russian anti-drone armor was largely absent. In addition to reducing speed and maneuverability, Russian crews have reported transmission failures with the added weight. Under pressure to continue fruitless offensive attempts, the assault suggests Russia's difficult in search of a more effective approach. (But) it is unclear whether adaptations will be sufficient to break through Ukraine's constantly evolving drone-centered defenses.

How Ukraine's Air Assault Troops Annihilated Russia's Tank Attack on Kramatorsk

The Russian command spent months planning an offensive to take Kramatorsk in Ukraine's Fortress Belt, only to have the assault annihilated by an even better planned Ukrainian counteroffensive which disrupted the Kremlin strategy with heavy losses. 

What continues to be evident in Russian initiatives is a lack of recognition that the Ukrainians are looking at the battlefield options the Russians have and then designing tactics for thwarting them with as much damage and as many casualties as possible. Why, after almost five years, the Kremlin refuses to recognize this, will be the stuff of great history some day. At present, it is the stuff of great Ukrainian intelligence, planning and performance. JL

David Axe reports in Trench Art and Clement Molin reports in X:

Two Russian combined arms armies spent months planning a powerful mechanized assault toward Kramatorsk. The attack failed on July 22nd. Ukrainian air assault brigade spoiled the attack in a complex operation beginning in April. Despite having been in the area for eight months, Russian forces had been unable to consolidate their positions; rivers complicated logistics, and it was impossible to concentrate large numbers of troops. The Ukrainians were able to advance stealthily behind enemy lines, reaching the Mokri Yali River. Counterattacking through gaps in Russian defenses 60 miles southwest of Kramatorsk, the paratroopers pinned down half of the 90th Division’s forces. So when the division attack in July, it attacked weakly. Ukrainian drones made quick work of the advancing T-72B3A tanks and other vehicles. This operation is taking place because the Russian army is unable to bring reinforcements to the front as anti-tank ditches prevent any Russian mechanized attempt at relief.

ChatGPT Now Refusing Requests For Text Mimicking Famous Authors

This suggests, if the pattern holds, that AI companies are starting to feel the legal heat about theft of intellectual property, which AI companies have, until now, blown off as the gripings of the jealous and insignificant. 

Combined with today's headline that Anthropic and OpenAI are calling on the US government to 'pause' the development of AI, possibly in response to OpenAI's agent hacking another company's files, about which OpenAI knew nothing for a week or more. This suggests that the companies are increasingly recognizing that they are going to need government assistance, financially and operationally but realize they are going to have to give something up in return. JL

Kyle Orland reports in ars technica:

OpenAI's ChatGPT is now refusing requests to generate text that directly mimics the style of famous authors. In refusing to directly copy the “exact style” of various authors, ChatGPT offered instead to capture an overall “feeling” by incorporating some of the features found in those authors’ work. That slight alteration could be legally important as OpenAI continues to fight a number of lawsuits brought by authors alleging large-scale copyright infringement by models trained on their work. One of those suits specifically cites ChatGPT’s “ability to generate text similar to that found in copyrighted textual materials.” In US copyright law an AI-generated stylistic imitation could become infringing if it becomes “substantially similar” to the work of the original author. 

Kupiansk Still A Kill Zone For Russia Due To Troop Fuel Shortages, Ukr Drones

Russian forces keep attacking Ukrainian positions at Kupiansk and, as a result, keep dying in droves. 

The high concentration of Ukrainian drones in addition to Russian troop and fuel shortages make it difficult for the Kremlin's units to do anything much beyond getting killed. JL

Kateryna Danishevka reports in RBC Ukraine and Ukranews reports:

Russian forces in the Kupiansk sector  are unable to conduct active offensive operations due to a lack of trained personnel. These units have also encountered severe fuel shortages, preventing them from making active use of motorcycles, all-terrain vehicles, and passenger vehicles. he high density of Ukrainian drone activity in this section of the front is preventing the Russian military from deploying a large number of its drones, including Lancet and Molniya.

The Ukraine and Iran Wars Are Merging - To Ukraine's Advantage

Iran has been supplying Russia with Shahed drones which have killed or severely wounded hundreds, if not thousands, of Ukrainians. And since the US and Israel attacked Iran (again) a few months ago, Russia has been providing Iran with satellite imagery and intelligence on US installations as well as air force and naval force locations and deployments. 

That Ukraine has now attacked an Iranian ship in the Sea of Azov, which connects Russia and Iran, signals that the two wars are now becoming more directly intertwined. This will work to Ukraine's advantage because it more securely ties Ukrainian interests with those of the US as the latter becomes increasingly frustrated with Iran's refusal to back down (even as the US president remains devoted to Putin and said he 'would ask Putin if the Russian help to Iran was true...'. It also reinforces Ukrainian relationships with the Gulf States to which it is already providing anti-drone weaponry. The merger, as it were, of the two wars makes Ukraine an even more important and uniquely qualified ally of those opposing Iran, while further undermining what Russian support remains in the US. Expect more Ukrainian attacks on Iranian assets to follow. JL

Eliana Silver reports in the DailyMailonline:

Russia's  war in Ukraine and the war in Iran could merge into one conflict after a Ukrainian strike on an Iranian ship heightened tensions between the two regions. Tehran condemned Ukraine for carrying out the strike on what it claims was a commercial vessel in the Caspian Sea, adding that the explosion killed one sailor and injured another. Ukrainian president Volodymyr Zelensky acknowledged the attack, adding that the Iranian vessel was being used to transport Russian 'military cargo.' He added  'Since the beginning of July, we have recorded active Russian satellite surveillance of the Gulf states and U.S. military facilities. There is a clear correlation between Russia's satellite imagery of these sites and Iranian strikes - both before the attacks, in preparation for them, and afterward, to assess the damage inflicted.'

Corporate America Has Suddenly Decided To Stop Blowing Money On AI

It was fun while it lasted, but who realistically believed that corporate customers for AI were going to blow their entire budgets on a new technology whose efficacy remained largely aspirational? 

Many AI company executives and venture investors may have embraced the Silicon Valley hype - 'breathing one's own exhaust' - as they say, but reality began to intrude earlier this spring as the costs went up and returns...remained elusive. What has emerged is the realization that some high end AI are good for specific tasks, but not necessarily for making your back office in Omaha a tad more productive. Different AI strokes - at different price points - for different folks is now becoming the norm. The looming question is, how are all those data center investments and related costs going to produce returns under this more prudent corporate spending scenario? JL

Angel Au-Yeung and colleagues report in the Wall Street Journal:

Companies across the country are coming around to a radical idea upending the industry powering the global economy: They don’t have to blow their budgets on AI. Fed up with ballooning costs, companies are using lower-priced models, including some built in China. In many cases, they are adding the new, cheaper models alongside OpenAI and Anthropic’s products, shopping a la carte for AI. Investors and executives say the era of cheaper AI is here to stay. Companies in Silicon Valley were the first to treat AI like any other product and shop for the best deal. The approach began spreading across industries in recent weeks. “There’s zero loyalty.” Strategies to lower AI costs include limiting access to top models for new hires and using the most advanced AI systems to plan how tasks will be completed before turning to cheaper models for the execution. “We work with all of them, figuring out solutions to improve performance and get much better cost.”